Professional Documents
Culture Documents
Plain English For Doctors and Other Medical Scientists 2017 PDF
Plain English For Doctors and Other Medical Scientists 2017 PDF
O S C A R L I N A R E S , MD
DAV I D DA LY
G E RT R U D E DA LY
1
iv
1
Oxford University Press is a department of the University of Oxford. It furthers
the University’s objective of excellence in research, scholarship, and education
by publishing worldwide. Oxford is a registered trade mark of Oxford University
Press in the UK and certain other countries.
This material is not intended to be, and should not be considered, a substitute for medical or
other professional advice. Treatment for the conditions described in this material is highly
dependent on the individual circumstances. And, while this material is designed to offer
accurate information with respect to the subject matter covered and to be current as of the
time it was written, research and knowledge about medical and health issues is constantly
evolving and dose schedules for medications are being revised continually, with new side effects
recognized and accounted for regularly. Readers must therefore always check the product
information and clinical procedures with the most up-to-d ate published product information
and data sheets provided by the manufacturers and the most recent codes of conduct and safety
regulation. The publisher and the authors make no representations or warranties to readers,
express or implied, as to the accuracy or completeness of this material. Without limiting
the foregoing, the publisher and the authors make no representations or warranties as to
the accuracy or efficacy of the drug dosages mentioned in the material. The authors and the
publisher do not accept, and expressly disclaim, any responsibility for any liability, loss or risk
that may be claimed or incurred as a consequence of the use and/or application of any of the
contents of this material.
We dedicate this book to the doctors and other medical scientists who write. They
strive to make the world a better place by writing about new ways to understand,
prevent, treat and cure disease.
vi
vii
CONTENTS
Preface xi
Acknowledgments xvii
Introduction 1
A . W HY B O T HE R TO W R I T E IN PL A IN E NGL I SH? 1
B. W H AT D O W E ME A N BY PL A IN E N G L I S H ? 4
E . NO T E S ON T HE EXE RC I SE S 10
viii Contents
F. D ON’ T BE A F R A ID TO S TA RT A SE N T E NC E W I T H A ND OR B U T 36
B. OMI T T HE NE E DL E S S OF 42
C . OMI T T HE NE E DL E S S T H AT 43
D. MINIMIZE F OR MS OF T O B E A ND T O HAV E 53
D. W R I T E IN T HE SINGUL A R 67
Contents ix
Glossary 201
Resources 205
Index 207
x
xi
PREFACE
xi
xii
xii Preface
After finishing medical school, I came back to the United States for residency.
Since I’d studied abroad, I had to take a test to qualify to work as a medical resi-
dent. Even though I knew the concepts, learning medical English was a big chal-
lenge. Luckily, both Spanish and English tend to use the same Latin and Greek
medical terms, which made things easier.
I did my fellowship in geriatric medicine at the University of Michigan, and
continued there for another 15 years as a visiting scholar. It was there I learned
medical writing through years of research and working with other doctors and
medical scientists.
Nobody ever actually taught me how to write about medicine. Instead, my
advisors told me to read articles in the journal you want to write for, study their
style, and imitate it. It was understood: You’re joining an exclusive club. You’ve
got to learn to write like members of the club.
This was hard advice to follow. Even after three years of training in internal
medicine, I had a tough time reading articles in geriatric neuroscience. They used
lots of long, formal sentences and impressive-sounding words. Many sentences
were in passive voice. Before I could understand an article, I had to study it care-
fully with Stedman’s Medical Dictionary at my side.
Whenever we wrote a scientific article, the writing process worked like this.
The lead author does the first draft, distributes it to co-authors for review, and
makes revisions. The senior author approves the final draft.
I always felt medical writing wasn’t as clear as it could be. It certainly wasn’t
reaching a wide audience. Only about 100 people in the world could understand
the articles we wrote. In fact, most of the best minds in medicine couldn’t under-
stand. A friend suggested I read Strunk and White’s book, The Elements of Style.
But, in those days, nobody in medicine seemed to know or care about writing in
plain English.
It was only when I started working with David a few years ago that I started to
see a better way to write. Together, we began to define medicus incomprehensibilis,
describe its symptoms, and explain how to treat it.
I, David, am a lawyer. I learned about plain English over the years from books,
people and lots of practice. I read Strunk and White’s The Elements of Style in
college. But nobody ever told me the ideas in that book might apply to math or
music, my two college majors.
While studying in Germany during my junior year, I learned what it was like
to be a “foreigner.” While I and the other foreign students were among the best
and brightest in every class, we always sat in the front row, straining to hear the
professor clearly. We often had minor troubles understanding ideas written in a
polite or formal German that differed from the German we spoke in everyday life.
In law school at the University of Michigan, I took the standard course on legal
research and writing, where we read Wydick’s book, Plain English for Lawyers.
I organized weekly German and Spanish conversation tables at the dining hall.
During my senior year, I taught the Introduction to American Law course for foreign
xiii
Preface xiii
legal scholars, which involved research and writing. These experiences helped me
see some of the difficulties non-natives face in reading and writing English.
Years later, I found myself at Chrysler working on international business deals.
There, I encountered many smart people, Americans and foreigners, who found
traditional contract language hard to understand.
Like most lawyers, I had learned traditional legal writing by imitating other
lawyers. When writing expert Bryan Garner came to Chrysler to put on semi-
nars on plain English for the legal staff, I was skeptical. But I tried his ideas and
found they worked. When he came back the next year, he pointed out my writing
sample as one written in almost perfect plain English. Somewhere along the way,
I also learned about Flesch Reading Ease scores. I found plain English was ideal for
international automotive deals, where people in different disciplines from all over
the world work together to build and sell cars.
After I had experience writing contracts in plain English, I wrote a few articles
for the Michigan Bar Journal. One called, “Why Bother to Write Contracts in Plain
English?”i has been cited on legal writing websites around the world. Another,
“Taming the Contract Clause from Hell: A Case Study,”ii got criticized by a reader
who thought my translation wasn’t quite right. In my response article, “The
Return of the Contract Clause from Hell,”iii I said he was right to call it a “trans-
lation.” The original was so convoluted, it might as well have been written in a
foreign language. Anything that bad was bound to be misunderstood. This whole
exchange became the subject of Prof. Joseph Kimble’s article, “The Great Myth
that Plain Language is not Precise,”iv and a chapter in his book, Lifting the Fog of
Legalese: Essays on Plain Language.v
We, Oscar and David, got the idea for this book a few years ago, when we
started working together to revise a paper on some of Oscar’s medical research.
At first, we had a clash in writing styles. But after dozens of shouting matches, we
worked everything out and the article got published. (One reviewer tried to turn
our plain-English into traditional medical writing.)
The final draft was far better than the original, and we traced the improvement
to a dozen or so principles. We thought other medical authors might benefit if we
could explain them. We started by writing down a list of rules for plain English
medical writing. One rule was, Decide what you need to say, then say it clearly and
concisely.vi This general rule applies all the time. Other rules were specific (e.g., talk
in terms of one doctor treating one patient), but only apply usually, often, or some-
times. Therefore, we thought it better to call these ideas, tips.
As we prepared to write, we checked other books on writing, found other
tips, and added them to our list. A few books we particularly liked were Anne
E. Greene’s Writing Science in Plain Englishvii and Joseph Williams’ Style: Toward
Clarity and Grace.viii As we added other people’s ideas to our own list of tips, we
noticed some ideas were more general and some, more specific. Sometimes,
different tips approached the same problem from a different angle. Some were
numerical (Keep the subject and verb together in the first seven or eight words), others
xiv
xiv Preface
grammatical or linguistic (Prefer the short word), some conceptual (Put the main
point first). We decided to include some of these different approaches, since some
readers might respond better to one approach or another.
I, Gertrude, graduated from the University of Michigan with a degree in
English. I am now a writer and editor. I run a blog about royal families with about
75% of my followers outside the USA.
Like Oscar and David, I’ve also had experience studying in another country.
During high school, I was an exchange student in Latvia. Some of my Latvian
classmates spoke English better than others, but it always came out as a Latvian
kind of English. They were obviously thinking in Latvian and translating their
thoughts. They tended to use shorter sentences, simpler grammar, and particular
words and phrases that translated well from Latvian into English. As they did
this, I found myself talking more like them so we could communicate better.
Later on, when I studied medieval, renaissance and early-modern literature at
the University of Michigan, I learned how English has changed over time. Jane
Austin, Shakespeare, Chaucer and Beowulf: the further back you go, the harder it
is for a modern audience to understand. I also learned how English has changed as
it has spread throughout the world. In this book, we urge authors to write modern
English, keeping essential scientific terms, which help explain the science, but
without archaic, exotic or obscure language that might keep some readers from
understanding.
Preface xv
also experts and insiders in their field, fail to notice the missing steps. But what
is obvious to insiders may not be obvious to everybody else in the widest reason-
able audience. Often, outsiders interested in the topic could follow the train of
thought if only the expert would take care to spell out their steps of reasoning.
The tips in this book come with exercises, so you can practice what you learn.
We base most exercises on excerpts from six leading medical journals: Journal
of the American Medical Association, The Lancet, New England Journal of Medicine,
British Medical Journal, Mayo Clinic Proceedings, and American Family Physician. We
chose these journals since they represent today’s best “standard” medical writing.
Writing in plain English is a skill you must hone by practice. Our goal is to
point you in the right direction, give some general guidance and specific tips, and
then, encourage you to develop your own creativity and good judgment. For most
exercises, we conclude by asking you to make any other changes you can think of to
improve reading ease.
For the exercises in Chapters 1–6, we check reading ease scores for the original
excerpts and our revisions in the Exercise Key. In Chapter 7, we analyze these data
to show how reading ease improves.
The tips in this book are specific, easy to use, and they really work. Many are
simple (e.g., Use normal sentence length). Yet for each tip, we found examples in
journals where the authors did the opposite, and the editors and peer reviewers
let it pass.
Medical science writing is important. It is every medical writer’s duty to write
clearly and concisely. Learning to express complex ideas clearly and concisely is in
no way a remedial skill. Rather, it can only been seen as a sign of mastery.
Notes
i. Daly D, “Why Bother to Write Contracts in Plain English?” Mich Bar J, (August 1999): 850–851.
ii. Daly D, “Taming the Contract Clause from Hell: a Case Study,” Mich Bar J, (October 1999):
1155–1157.
iii. Daly D, “The Return of the Contract Clause from Hell,” Mich Bar J, (February 2000): 202–204.
iv. Kimble J, “The Great Myth That Plain Language Is Not Precise,” Scribes J Leg Writing
7 ( 1998–2000 ), 109.
v. Kimble J, Lifting the Fog of Legalese: Essays on Plain Language (Durham, NC: Carolina Academic
Press, 2006): 37–48.
vi. This rule came from one of Bryan Garner’s live seminars.
vii. Greene A, Writing Science in Plain English, (Chicago: University of Chicago Press, 2013).
viii. Williams J, Style: Lessons in Clarity and Grace, 9th ed. (New York: Pearson Longman, 2007).
xvi
xvii
ACKNOWLEDGMENTS
xvii
xviii
1
Introduction
Since writing is the only means the medical profession has of universally
disseminating knowledge concerning new therapeutic concepts, medical
discoveries, or clinical experience, it is the moral obligation of every
physician who has made an original scientific observation or has for-
mulated from his own experience a new medical theory, to publish
it for the information of his colleagues, and the ultimate benefit of
mankind.—Selma DeBakey1
1 . I T HE LP S S PR E A D N EW ME DIC A L K NO W L E DG E
When doctors share ideas about theory and practice, it helps spread medical
knowledge. New medical discoveries prevent illness, relieve suffering, find cures,
and extend life. Difficult-to-understand writing slows down this process; clear,
concise writing speeds it up.
2 . I T HE LP S T E AC H T HE PROF E S SION
The Hippocratic Oath is a tradition of the medical profession; many doctors take it
when they graduate from medical school. As part of the oath, they pledge to teach
others the profession “according to their ability and judgment.” This implies they’ll
do what they can to make medicine understandable to others.
3 . I T S HO W S R E SPE CT F OR T HE R E A DE R
Doctors are busy professionals. When you write in plain English, it shows respect.
Your reader will read faster, understand better, and remember longer. Research in
1
2
other fields shows 80% of readers prefer plain English.2 Even if they can under-
stand an article written in a traditional style, doctors are human too. Their brains
work the same way as everybody else’s. The same factors that influence reading
ease for everybody else also influence reading ease for doctors.
4. I T S AV E S R E A DIN G T IME
How much time does the average doctor spend reading medical journals? Research
published in the Journal of the Medical Library Association reports the average
pediatrician spends 118 hours a year.3 The Journal of General Internal Medicine
reports the average internist spends 228 hours a year.4
What is the potential savings in reading time? Since there is little research on
plain English medical writing, we can only make an educated guess based on expe-
rience in other fields. Writing expert Robert Eagleston thinks writing in plain
English may cut reading time by 30% to 50%.5 Joseph Kimble tested traditional
and plain English contracts on various groups of readers and found that plain
English cut reading time between 4.7% and 19.7% while improving comprehen-
sion.6 Considering how much doctors read and how much time plain English
saves, it seems likely that, if all medical journals were written in plain English, it
could save the average doctor a week or two per year.
Writing well may take extra work, but even if it does, it’s worth it. If your arti-
cle is published and circulated to thousands of readers, any extra time you spend
to write more clearly and concisely is small compared to the total time you save
for your readers.
When you consider the widest reasonable audience, and write in a style suitable for
them, it promotes free and efficient exchange of new medical knowledge. Ideally,
a medical journal article should be accessible to anybody interested in the sub-
ject matter, whether or not they are an insider in a field. This may include a doc-
tor or scientist working in the same or a related specialty, a student, or a nurse.
It includes regular journal subscribers and those who search for articles on the
internet.
English is the global language of science, but many people who read English-
language medical journals are not native speakers. In fact, according to English-
language expert David Chrystal, non-native English speakers outnumber native
speakers 3:1.7 The widest reasonable audience includes those living or educated in an
English-speaking country or elsewhere, whether a native speaker of English or not.
In Appendix 1, we present a survey of English speakers around the world. Some
of the conclusions we reached in putting it together surprised us. For example, did
3
Introduction 3
you know there are now 23 countries where more than 10 million people speak
English? Did you know there are more people who speak English in Asia than in
North America? And did you know Germany is now the seventh largest English-
speaking country in the world with more than 50 million English speakers? Many
German doctors read English-language medical journals. They may have a first-
rate medical education but only read English at a USA high school level. Given
that English-language medical journals are read by doctors throughout the world,
it makes good sense to write complex science in plain English.
6 . I T HE LP S PL E A D F OR A C AU SE
Medical science articles advocate for people who are poor, sick or oppressed. If you
don’t make your point clearly, you’re not helping anybody. Writing in an inflated
formal style sends the message there’s no urgent problem. It’s just business as usual
for those of us who work in a hospital, university or research center. Writing in plain
English helps send the message that a problem is important and urgent.
7 . I T HE LP S H UM A N IZE YOUR W R I T IN G
Writing in plain English sounds more natural, closer to the way people speak in
everyday life. It sounds professional, with careful attention to the science, but less
formal and bureaucratic. This means your writing sounds more human and puts
less of a burden on your reader.
Your research and ideas deserve to be presented clearly and concisely. Good writ-
ing helps build your reputation and benefits your career. More people will read
your work and want to work with you.
Writing in plain English helps you overcome editorial blindness, that feeling you
get when you work on an article so long you miss problems a reader with a fresh
eye would see. The tips in this book help you look at your writing from a fresh
perspective.
1 0 . I T S AV E S T IME A N D IMPROV E S C ON T E N T
Most medical journal articles involve multiple authors and go through peer
review. This involves many people and many steps. Getting the first step right, by
making sure your draft is clear and concise, saves time and effort at each later step.
Let’s look at the process from the lead author’s point of view. You start by writ-
ing the first draft as clearly and concisely as you can. Next, your co-authors review
the draft. Since the draft is clear and concise, they focus more on content and less
on form. They read faster, understand better, and give better comments. They
become more personally invested in the work, and team morale improves.
Once you get comments, you incorporate them into a revised draft. Since the
original was clear and concise, this revision goes quickly. Since you got better
comments from your co-authors, the content gets better. Since you continue to
check reading ease as you revise, the revised draft becomes even clearer and more
concise.
Now, your co-authors review the second draft, which reflects everybody’s com-
ments. They see the improved content and this lifts team morale further. They
become more invested in the research effort. You and your co-authors continue to
review and revise as long as you need.
Once you’re satisfied with the article, you submit it to a journal. If the journal
likes it, they assign it for editorial and peer review. Just as with co-author review,
this review goes faster and generates better comments.
Writing a first draft in plain English may take longer, but, after that, each step of
review and revising goes faster and helps improve clarity, conciseness and content.
As a result, you end up with a better paper, possibly with no more time and effort.
You can’t learn to write in plain English just by reading a book. You must put
your own pen to paper.8 This is why we included exercises for each chapter.
1. What types of people might want to read an article about modeling how a
kidney filters blood? In other words, who is the widest reasonable audience?
2. What changes would you suggest to make the article clearer for different peo-
ple in the widest reasonable audience? (E.g., a mathematician might need a
better explanation about kidney anatomy.)
Introduction 5
dumbing down medical science, but this is not so. Rather, it involves sharpening
up the science to make it clearer and more accessible to the widest reasonable
audience.
By content, we mean essential scientific content, those important scientific
ideas an author must include in their article. What content is non-essential?
Sometimes, an author loads down a sentence with asides and parentheticals that
only loosely relate to the main idea. Only an author can judge what content is, or
isn’t, essential. An author should never sacrifice essential content to make their
article easy to read, but they might cut non-essential content.
Writing is clear when the narrative uses words and concepts familiar to the
reader. Ideally, a reader can understand and vividly imagine the article on first
reading without having to study it. The reader remembers each key idea.
Writing is concise when it demands as little of the reader’s mental energy as
possible. This usually means short while still clear. Good writing involves trade
offs. A few short words may convey the message more vividly than one long but
lifeless word. Writing concisely means cutting an unnecessary word, but cutting
too many words may make the message cryptic and harder to understand. For
example, a math equation is very concise, but a reader might understand the same
idea better if the author explained it in words first.
Figure I-1 presents a diagram that shows how we think about plain English.
Within the universe of all the possible ways you might write a medical journal
article, plain English represents the intersection of three ideas: (1) essential sci-
entific content (2) presented clearly and (3) concisely. This diagram represents
plain English—what we are trying to achieve. We devote the rest of this book to
explaining how to achieve it.
Essential
scientific
content
Wordy Cryptic
Plain
English
Dumbed
Clear Concise
down
1. Count the number of words in each sentence and compute the average sen-
tence length. Which excerpt uses the longest sentences? The shortest?
2. Which excerpt do you find easiest to read? The hardest?
3. Judging from these excerpts, do you see any link between reading ease and
sentence length?
Introduction 7
a problem. But the more symptoms a piece of writing has, the worse the case of
medicus incomprehensibilis.
“ T I P S ” —N O T RUL E S
Everybody who writes a book about writing style gives a list of rules. In develop-
ing our list, we checked other books and articles and tried to include all the best
ideas that apply to medical writing. (See Resources.)
8
Like medicine, writing is both an art and a science. It involves many choices
about how to say things. There is only one rule for plain English medical writ-
ing: Decide what you need to say, then say it clearly and concisely. This rule applies all
the time.
We call the other suggestions in this book tips. Most are specific and easy to
apply. We expect an author to use good judgment about when a tip may prove
helpful. We offer each tip with the provisos: “when it helps,” “when it works,”
“when it makes sense,” and “if it isn’t confusing, distracting, or vulgar.” For exam-
ple, one tip is, Prefer active voice. Many medical writers use passive voice too much
and their reading ease and vividness suffer as a result. But, if you use passive voice
sparingly, it can work well.
Think of each tip like a tool in your toolbox of writing skills. Like any tool,
not every tip helps in every situation. A hammer doesn’t help much to fix a
clogged sink. If a tool doesn’t work in a particular situation, that doesn’t mean
you throw it away. Instead, you keep it in your toolbox for when it serves a use-
ful purpose.
PL A I N E NG L I S H W R I T IN G T IP S HE LP IMPROV E A L L K I NDS
OF W R I T I N G
Many medical journal articles suffer from medicus incomprehensibilis; some are
worse than others. Some are almost unreadable. Some are tedious. Some are
readable, but take study. Still others are easy to read, but wordy. In this book, we
address each of these issues. The tips we provide can help make poor writing good,
and good writing better. Don’t stop improving just because you feel your writing
is good enough.
S M A L L C H A N G E S A DD UP
Some of the writing tips we present in this book may seem trivial, but many
small changes quickly add up to a big improvement in reading ease and clarity.
Each individual tip may only improve your article’s reading ease score by a few
points, make one idea a little more vivid, or improve the flow of logic in a small
way. But by making many small changes, you may see your reading ease score
jump, a hazy idea become clear, or a weak chain of logical reasoning become
strong.
K A I Z E N (改 善)
Introduction 9
eliminate waste. Kaizen is a daily process, which goes beyond improving produc-
tivity. It is also a process that, when done correctly, humanizes the workplace,
eliminates overly hard work (muri), and teaches people how to spot and elimi-
nate waste in business processes. While kaizen usually delivers small improve-
ments, the culture of continual small improvements yields large productivity
improvements.9
C H A L L E N G I NG C ON V E N T ION
Some of our tips differ from traditional notions about “proper” medical writ-
ing. We challenge any convention that serves no scientific purpose, yet makes
an article needlessly complex or abstract. If you ever feel one of our tips violates
a convention of the profession, ask yourself, Is this convention based on a scien-
tific reason, or a social reason? If the reason is scientific, by all means, follow the
convention.
But if the convention is based on a social reason, ask yourself, How impor-
tant is that social reason? Is it more important than respecting a reader’s time? Is it
more important than making an idea accessible to the widest reasonable audience?
The benefits of plain-English outweigh any superficial notion about “proper”
writing.
You might ask, Aren’t medical writing habits that contribute to medicus incompre-
hensibilis of long standing? Not so much as some people think. Many medical writ-
ers of the past wrote their scientific works clearly. For example, Edward Jenner’s
1798 paper, “An Inquiry into the Causes and Effects of the Variolae Vacciniae,
or Cow-Pox,” which first coined the term vaccination,10 was written for a wide
audience and does not overuse difficult scientific language. Watson and Crick’s
1953 article, “Molecular Structure of Nucleic Acids: A Structure for Deoxyribose
Nucleic Acid,” which first proposed the double-helix structure of DNA, is short
and readable.11
S TAG E S OF G R IEF
Conclusion
If you’ve read this far, you’ve already taken the biggest step toward writing better.
You accept that it’s possible to write about medicine in plain English and that it’s
worthwhile. Most importantly, you’re willing to try.
Notes
1. DeBakey S, “Suggestions on Preparation of Medical Papers,” JAMA 155, no. 18 (1954): 1573.
2. Kimble, Lifting the Fog of, 3–13 (see Preface, n. 5).
3. Tenopir C, et al. “Journal Reading Patterns and Preferences of Pediatricians,” J Med Libr Assoc 95,
no. 1 (2007), under “Background,” http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1773049/.
4. Saint S, et al. “Journal Reading Habits of Internists,” J Gen Intern Med 15, no. 12 (December
2000), under “Reading Habits,” http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1495716/.
11
Introduction 11
concept
1
TAKE CHARGE OF YOUR
READING EASE SCORE
Everything that can be thought at all can be thought clearly. Everything
that can be said can be said clearly.—Ludwig Wittgenstein1
In the first three chapters, we discuss how you can improve your writing by
improving your reading ease scores. We address reading ease first, since it repre-
sents the “low hanging fruit” of writing in plain English.
1
Flesch, “How to Write in,” (see Concept 1, n. 4).
should we expect for plain-English medical writing? Because medical writing uses
Latin or Greek names and other scientific terms, we think a good reading ease
score tends to be a bit lower (about 46–70), and the grade level a bit higher (about
7–11). We show in Chapter 7 how we came up with these ranges.6
The Flesch Reading Ease and Flesch–Kincaid Grade Level tests give a good gen-
eral idea of how difficult it is to read a passage of text. They are also quick and
easy to calculate. Chances are your computer’s spell checker can calculate them
for you.
We use these tests the way a doctor uses a blood pressure cuff, to get some
useful information quickly at low cost. We often use them in a way that is “off
label,” to check just one sentence or paragraph rather than a whole article. But,
the shorter the block of text, the more unreliable the score may be in reflect-
ing actual reading ease. Some things are probably not so hard for a doctor to
read as the score might suggest (e.g., names of body parts, medicines, surgical
procedures).
W HE R E C A N YO U FIN D T HE SE T E S T S?
The Flesch Reading Ease and Flesch-Kincaid Grade Level tests come as a standard
feature of the MS Word spelling and grammar checker. They are also bundled with
other popular word processing programs and services. (You may need to enable
style checking on your word processing program. Check the program’s website
for user help.) Different versions may provide slightly different scores. If your
word processing program doesn’t have these tests, you can find them free online
(https://readability-score.com).
15
0.0 μT μPE
Flesch Reading Ease (FRE) score
Figure C1-1 Plain English can materially improve a journal’s reading ease.
16
article in the American Family Physician, a journal read by 190,000 family practice
doctors.
We understand this sentence, but it’s not as easy to read as it might be. An occa-
sional sentence like this doesn’t pose a problem for the average medical journal
reader. But when every sentence sounds like this, an article soon becomes tedious
and fails to serve the widest reasonable audience.
The wseg score gives us a few hints about this sentence. First, it is a little
long: one sentence of 27 words. The reading ease score is 0.0. The grade level is
20.7, which may not pose a problem for a doctor educated in English. But it might
for a doctor educated in another language.
What makes this sentence challenging to read—the science or the writing
style? Certainly, understanding the term, Noonan syndrome, requires some medi-
cal knowledge. But, in this case, the whole article is about Noonan syndrome; the
reader will likely learn what it involves. A reader can also tell Noonan syndrome is
a medical condition. Thus, the science is a not the cause for this sentence’s low
reading ease.
As we see it, the low reading ease springs from commonly overused writing
habits:
• Many words come between the subject (referral) and the verb (may)
• A long sentence—27 words
• A long dependent clause in the middle (including determining the appropriate-
ness and sequence of genetic counseling)
• Abstract nouns: referral, assistance, diagnosis, management, appropriateness and
sequence
• Slight redundancy—determining the sequence of genetic testing is part of
determining appropriateness
If we try to revise the example to keep the same scientific content, but avoid or
minimize these writing habits, we might end up with something like the revision
in Table C1-2.
In this revision, the number of words is the same, but since the words are
shorter, it takes up less space on the page. Since it uses two sentences, the aver-
age sentence length is one-half. The Flesch Reading Ease score jumps from 0.0
to 61.5. The grade level goes down from 20.7 to 8.0. Clearly, if every sentence
17
Original Revised
Text Referral to a clinical geneticist It may help to refer a patient with
for assistance in the Noonan syndrome to a clinical
diagnosis and management geneticist. They can help diagnose
of Noonan syndrome, and manage it and help decide on
including determining the genetic tests. (wseg = 27/13.5/
appropriateness and sequence 61.5/8.0)
of genetic testing, may be
helpful.
(wseg = 27/27/0.0/20.7)
Words 27 27
Average words
27 13.5
per sentence
Reading ease
0.0 61.5
score
Grade level 20.7 8.0
in every article could be revised like this, a journal could serve a much wider
audience.
Is this the best possible revision? Maybe you can think of something better. Or
maybe you would have made some different revising decisions than we did and
achieved a different wseg score. The point is, when you focus on reading ease, you
can often make a big improvement.
1. Read the following excerpt out loud and underline each word with three syl-
lables or more (but skip expanded and produces, which are two-syllable words
that add common endings).
2. For each word you underlined, can you think of a shorter word or a few short
words that mean close to the same thing? Or else, do you consider it an essen-
tial scientific term, not easy to replace?
3. Revise the excerpt to break up the long sentences into shorter ones. Keep any
essential scientific term, but use shorter words where you can. Do you find
your revision easier to read than the original?
18
Conclusion
The first concept is Take charge of your reading ease score. A reading ease score is
based on two factors: average sentence length and average word length. In the
next three chapters, we give tips for improving reading ease scores:
Notes
1. Williams, Style: Lessons in Clarity, 3 (see Preface, n. 8).
2. For the exercises in this book, negative scores are truncated to zero. This is a standard feature
for the MS word style checker. We present an analysis of the wseg scores for excerpts used in
Chapters 1–6 exercises and our revisions in Chapter 7. For this analysis, we calculated negative
scores manually.
3. Wikipedia, s.v. “Flesch- Kincaid Readability Tests,” https://en.wikipedia.org/wiki/Flesch
%E2%80%93Kincaid_readability_tests (accessed February 14, 2016).
4. Flesch R, “How to Write in Plain English,” University of Canterbury, http://www.mang.canter-
bury.ac.nz/writing_guide/writing/flesch.shtml (accessed February 12, 2016).
5. Ibid.
6. In general, these ranges represent one standard deviation around the mean of our model revi-
sions. See our analysis in Chapter 7.
7. Bhambhani V, Muenke M, “Noonan Syndrome,” Am Fam Phys 89, no. 1 (January 2014): 40.
8. Carter R, “B Cells in Health and Disease,” Mayo Clinic Proc 81, no. 3 (March 2006), under “B Cells
and Disease States,” http://www.mayoclinicproceedings.org/article/S0025-6196(11)61466-3/
fulltext.
19
CHAPTER 1
Use normal sentence length
Modern English, especially written English, is full of bad habits which
spread by imitation and which can be avoided if one is willing to take the
necessary trouble.—George Orwell, Politics and the English Language1
What makes traditional medical writing hard to read? One thing is long sentences.
Controlling sentence length helps improve your reading ease score. Consider the
following excerpt from Watson and Crick’s 1953 article in Nature, which first
described the double-helix structure of DNA:
Even though this excerpt describes a complex topic in molecular biology, the aver-
age sentence length is less than 20 words. The reading ease score is 52.7 and the
grade level is 10.0.
19
20
control. We think this means about 15 words average and 25 words maximum. Of
course, when we say, 15 words average, we mean some sentences are longer and
some shorter.
There are exceptions. For example, you might write a longer sentence if your
sentence ends in a list, or if you quote the long title of a study. Generally, if a
sentence runs more than 25 words, it covers more than one idea and you should
consider splitting it into two or more sentences. You’ll find it’s often easy to split
up a long sentence to make shorter ones.
Writing style experts recommend mixing long and short sentences.3 Many
short sentences, one after another, can seem choppy and distract the reader.
Many long sentences, one after another, become tedious to read.
N O S C I E N T I F IC R E A S ON F OR L ON G SE N T E NC E S
The science of reading ease tells us long sentences tend to be hard to read. In this
respect, there is a good scientific reason to avoid long sentences. A complex idea
may take longer to explain, but sentences of moderate length will serve just as
well or better.
C OMMON W I SD OM AB OU T SE N T E N C E L E NGT H M AY NO T
A PPLY TO M E DIC A L W R I T IN G
Because of their unique subject matter, terms and audience, a medical author
should strive for an average sentence length of about 15 words and a maximum
sentence length of about 25 words.
Exercise 1.B. Keep the subject and verb close together in the first seven or
eight words
For each sentence:
1. Read it out loud and underline the subject once and the verb twice. Count
the number of words that come between them. Do the subject and verb come
within the first eight words?
2. Revise to make new shorter sentences. Each new sentence should have its sub-
ject and verb close together within the first seven or eight words. Make any
other changes you can think of to improve reading ease.
It also makes sense to state your premises before you state a conclusion.
But when a writer starts a long sentence with a long dependent clause, it can
tax the reader’s powers of concentration. For example, consider the sentence
below where the main point starts 18 words into the sentence (in italics). We can
improve reading ease by breaking up the long sentence, putting the main point
first, and minimizing other long words (Table 1-1).
Sometimes a long sentence contains several key points. In such a case, it helps
to split up a long sentence into shorter sentences, with each key point near the
start of its own sentence.
If you want to help a busy reader understand your ideas, put the main point
first, and then give commentary, detail or support.
24
Original Revised
Although reducing heart rate For a patient with sepsis, a slow heart rate
will decrease myocardial oxygen may cut cardiac output and tissue perfusion.
consumption and will improve diastolic For a patient without sepsis, slowing the
function and coronary perfusion, heart rate slows the heart’s oxygen use. It
for patients with sepsis, an inadequate also helps diastolic function and coronary
chronotropic response may potentially perfusion. (wseg = 38/12.6/55.9/8.6)
negatively affect cardiac output and tissue
perfusion.i (wseg = 34/34.0/0.0/24.3)
i
Morelli A, et al. “Effect of Heart Rate Control with Esmolol on Hemodynamic and Clinical
Outcomes in Patients with Septic Shock,” JAMA 310, no. 16 (2013): 1689.
Exercise 1.C. Put the main point first and then give commentary, detail or support
For each sentence:
1. Read it out loud and underline the main point or each key point.
2. Count the number of words that come before the main point or each key point.
3. Does the main point come near the start of the sentence? If not, revise to
break up any long sentence. Put the main point first or put each key point
near the start of its own sentence. Make any other changes you can think of to
improve reading ease.
4. The participation rate was lower than we expected when we did the
power calculations; however, taking into account the fact that more
people than expected had an increased risk and received counselling
and that not even a trend to a reduction in ischaemic heart disease
was observed, we doubt that a participation rate of 70% would have
made any difference.26
5. Recommended CRC screening strategies fall in 2 broad catego-
ries: stool tests that primarily detect cancer, which include detection
of occult blood or exfoliated DNA, and structural tests, such as flexible
sigmoidoscopy, colonoscopy, and computed tomographic colonography,
which are effective in detecting both cancer and premalignant lesions.27
6. According to the AUA, the presence of three or more red blood cells
on a single, properly collected, noncontaminated urinalysis without
evidence of infection is considered clinically significant microscopic
hematuria.28
Conclusion
Use normal sentence length to make your writing more readable. There is no sci-
entific reason to write a long sentence. On the contrary, the science of reading
ease indicates you should keep your sentence length moderate. We suggest the
right length for medical writing is about 15 words average, with a maximum of
about 25 words. Keep the subject and verb together in the first seven or eight
words. Put the main point first.
In the next chapter, we consider word length and its effect on reading ease.
Notes
1. Orwell G, “Politics and the English Language,” Horizon (April 1946). Available online, http://
www.orwell.ru/library/essays/politics/english/e_polit/.
2. Watson, “Molecular Structure of Nucleic,” Nature (see Intro, n. 11).
3. See for example Wydick, Plain English for Lawyers, 36 (see Intro, n. 8).
4. See for example Kimble, Lifting the Fog of, 71 (see Preface, n. 5); Wydick, Plain English for Lawyers,
36 (see Intro, n. 8).
5. See for example Brooks L, Building Great Sentences: Exploring the Writer’s Craft, (Chantilly,
VA: The Great Courses, 2008).
6. Cutts M, Oxford Guide to Plain English, 3rd ed. (Oxford: Oxford University Press, 2009), 2.
7. Alfandre D, Henning-Schumann J, “What is Wrong with Discharges Against Medical Advice (and
How to Fix Them),” JAMA 310, no. 22 (2013): 2393.
8. Mannino D, Bust A, “Global Burden of COPD: Risk Factors, Prevalence, and Future Trends,”
Lancet 370 (September 2007): 765.
26
CHAPTER 2
Prefer the short word
If you want to influence your reader with your ideas, resist the tempta-
tion to use long Latin-or French-based words where shorter ones will
do. Your message will be clearer and have much more impact.—Anne
Greene, Writing Science in Plain English1
This chapter talks about how word length affects reading ease. If you want to write
about medicine clearly and concisely, prefer the short word. Never use a big word
just because you can. Use a big word when it helps you explain the science clearly
and concisely. Impress people with your good ideas, not your big words.
27
28
3. Doctors and other medical scientists use the term consistently (i.e., exclu-
sively), and
4. It’s easy to look up in a standard reference.
Using these same tests, pulmonary does not qualify as an essential scientific
term, since the word lung often serves just as well, and doctors use both terms.
1. Read it out loud and underline each long word (i.e., three syllables or more, not
counting common word endings).
2. Double underline each word you consider an essential scientific term.
3. For any word you underlined only once, try to substitute a shorter word or
paraphrase using a few shorter words. Make any other changes you can think
of to improve reading ease.
5. The 2 objectives of the study were (1) to examine the association
between physical activity and dietary behavior and (2) to exam-
ine the potential combined effect of physical activity and dietary
behavior on biological (eg, total cholesterol) and health (eg, waist
circumference) markers.6
6. Assuming vasal disruption and occlusion have been adequately
achieved during surgery, and assuming the patient adheres to using
another contraceptive method while awaiting confirmation of sterility,
true causes of vasectomy failure include recanalization (early and late)
and, more rarely, aberrant anatomy (e.g., the presence of a third vas).7
word plus a prefix. Medical writing uses many compounds formed from Latin or
Greek root words. Try to handle compounds carefully, since:
• The longer the word, the lower the reading ease, and the less familiar it is
likely to be.
• Any gap between how a word is written and how it is pronounced may slow
down a reader.
• A reader may fail to see a word is a compound and puzzle over its meaning.
• An uncommon compound may be difficult to look up.
An occasional long compound may help explain the science clearly. But if you try
to use every big word you know, it will lead to a severe case of medicus incompre-
hensibilis. You will lose many readers and waste other readers’ time.
There are three main types of compounds: the open compound, the closed compound,
and the hyphenated compound. With the open compound, words work together but
are written as separate words. Examples include: student nurse, 50 percent, and ref-
erence book. Since the words aren’t connected, you might not even think of them
as compounds.
With the closed compound, words, or a word plus a prefix, are written as one
word. Examples include: multicell, hyperadrenergic, vasomotor and sinoatrial. For
some closed compounds, the meaning and syllable stress change. (For example,
compare the compound, straightforward with the separate words, straight and
forward. The compound has a different syllable stress and carries a slightly dif-
ferent meaning.) There is a general trend towards combining words to make
new closed compounds (e.g., on-line becomes online).8 In medical writing, this
same trend can sometimes lead to ungainly long words that add to medicus
incomprehensibilis.
With a hyphenated compound, words, or a word plus a prefix, are written
together but separated by a hyphen. Examples are: pre-menstrual, cost-effective,
self-reported. With the hyphenated compound, the component words are pro-
nounced as separate words.
S TA N DA R D U S AG E F OR C OMP OUN DS
For best reading ease, medical writing should follow standard usage regarding
compounds and hyphenation. These rules are detailed and complex. For example,
The Chicago Manual of Style contains 10 pages on compound words and hyphen-
ation. Focusing on a few key guidelines can help a medical writer reduce medicus
incomprehensibilis:
31
• Use caution in writing a word as a closed compound, unless that form is widely
accepted and pronunciation and reading ease are not at stake.
• A well-placed hyphen can often make for easier reading.
• Words that can be misread should be hyphenated.
• Where no ambiguity would result, a hyphen is not mandatory. It may be better
to write the word as an open compound.9
S T R AT E G I E S F OR M A N AG IN G C OMPOUNDS
Each of these examples might confuse a reader on first reading, if only for an
instant. For example, somebody might easily misread the word postherpetic as
POS-therpetic. Writing this word as a hyphenated compound, post-herpetic helps
the reader better see the units of meaning.
Some of the same considerations apply to prefixes. Consider the words decon-
taminate or deactivate. We pronounce them as if they were written as D-contaminate
and D-activate. (We don’t say DECK-kon-ta-mi-nate or DEAK-ti-vate.) For an inter-
national medical audience, we think it would be better to write de-contaminate and
de-activate.
W HE N WO ULDN ’ T W E HYPHE N AT E?
Guided by these principles, we would not hyphenate prefix, react, prototype, ultra-
sound or phenotype. There may be other cases where you choose not to hyphenate
a compound word. But you should always try to write a compound word as clearly
as you can to help reduce medicus incomprehensibilis.
Some journals may have a house style for writing compounds that differs from
the advice we give here. If you take care to write compounds clearly, but an editor
or reviewer insists on using their house style, by all means, defer to their judgment.
Exercise 2.C. Write a compound word to promote reading ease and show how
you pronounce it
While you’re in Stockholm to accept your Nobel Prize, you dine with other prize
winners who speak English, but not as well as you. One of them asks for your help
to pronounce the words underlined in the examples.
33
1. For each word underlined, either add a hyphen to help show the word’s mean-
ing or pronunciation, or tell why you think adding a hyphen wouldn’t be
helpful.
2. Revise to make any other changes you can think of to improve reading ease.
Which endings do you need to convey your meaning clearly and concisely?
Your judgment may be as good as ours or better. Whenever you can reasonably
leave off a word ending, it helps improve reading ease.
1. Read the sentence out loud. Do you think it would work to delete the end-
ing and just use the root word? Or else, to use a simpler word? Why or
why not?
2. Revise the sentence, making any other changes you can think of to improve
reading ease.
You may want to use a short, familiar noun string to help name a complex con-
cept in just a few words. Otherwise, avoid long strings that are not already well
known.26 Avoid creating a new noun string.27
1. The study by Vinden et al provides direct evidence from patient
outcomes that operating the night before is not associated with
increased complications for elective laparoscopic cholecystectomies
performed the following day.28
2. Performance is assessed in terms of rigorous out-of-sample predic-
tive validity testing based on the root-mean-squared error of the log of
the age-specific death rates, the percentage of time that trend is accu-
rately predicted, and the coverage of the uncertainty intervals (UIs).29
3. By targeting VKOR, the post-translational modification of the
vitamin K-dependent blood-coagulation proteins is impaired.30
6. Tick paralysis, which results from gravid female bites, is a toxin-
mediated ascending paralysis that generally resolves after tick
removal.33
1. Read it out loud and underline the long conjunction that starts the last
sentence.
2. Try replacing that long conjunction with and, but, or some other short word,
and read it out loud again. How does it sound to you?
3. Revise to make any other changes you can think of to improve reading ease.
Conclusion
The two easiest things you can do to make your writing more readable are to use
normal sentence length and prefer the short word. Just as there is no scientific
39
reason to write a long sentence, there is no scientific reason to use many non-
essential long words. On the contrary, the science of reading ease indicates you
should prefer the shortest word that does the job. When you can master these two
lessons, you will have gone a long way toward improving reading ease.
Notes
1. Greene, Writing Science in Plain, 30 (see Preface, n. 7).
2. Feres F, et al. “Three vs. Twelve Months of Dual Antiplatelet Therapy after Zotarolimus-eluting
Stents: The OPTIMIZE Randomized Trial,” JAMA 310, no. 23 (2013): 1517.
3. Wang, “Age-specific and Sex-specific Mortality,” 2071–2072 (see chap. 1, n. 24).
4. Weinshilbourn, “Inheritance and Drug Response,” 529 (see chap. 1, n. 17).
5. Mallett, “Systematic Reviews of Diagnostic,” under “Introduction,” (see chap. 1, n. 18).
6. Loprinzi P, Smit E, Mahoney S, “Physical Activity and Dietary Behavior in US Adults and Their
Combined Influence on Health,” Mayo Clinic Proc 89, no. 2 (February 2014): 190.
7. Rayala B, Viera A, “Common Questions about Vasectomy,” Am Fam Phys 88, no.
11(2013): 759.
8. Chicago Manual of Style, 15th ed., s.v. § 7.84.
9. Ibid. §§ 7.84–7.90.
10. Adren [L. ad, to + ren, kidney] + [G. Ergon, work]; Stedman’s Medical Dictionary, ed. 28, s.vv.
“Hyper,” “Adrenergic.”
11. Morelli, “Effect of Heart Rate,” 1689 (see chap. 1, Table 1-1).
12. Mannino, “Global Burden of COPD,” 767 (see chap. 1, n. 8).
13. Pirmohamed M, et al. “A Randomized Trial of Genotype-Guided Dosing of Warfarin,” N Engl J
Med 369, no. 24 (2013): 2295.
14. Bertoia M, et al. “Dietary Flavonoid Intake and Weight Maintenance: Three Prospective
Cohorts of 124,086 US Men and Women Followed for up to 24 Years,” BMJ 352, no. i17
(2016), under “Introduction,” http://dx.doi.org/10.1136/bmj.i17.
15. Singh B, et al. “Endovascular Therapy for Acute Ischemic Stroke: A Systematic Review and
Meta-Analysis,” Mayo Clinic Proc 88, no. 10 (2013): 1064.
16. Zimmerman T, “Common Questions about Barrett’s Esophagus,” Am Fam Phys 89, no. 2
(2014): 96.
17. Vigen R, et al. “Association of Testosterone Therapy with Mortality, Myocardial Infarction, and
Stroke in Men with Low Testosterone Levels,” JAMA 310, no. 17 (2013): 1834.
18. Mannino, “Global Burden of COPD,” 769 (see chap. 1, n. 8).
19. Weinshilbourn, “Inheritance and Drug Response,” 529 (see chap. 1, n. 17).
20. Miller A, et al. “Twenty-Five Year Follow-up for Cancer Incidence and Mortality of the
Canadian National Breast Screening Study: Randomized Screening Trials,” BMJ 348 (2014)
under “Introduction,” http://www.bmj.com/content/348/bmg.g366.
21. Dunlay S, Pereira N, Kushawaha S, “Contemporary Strategies in the Diagnosis and Management
of Heart Failure,” Mayo Clinic Proc 89, no. 5 (May 2014): 663.
22. Caylor T, Perkins A, “Recognition and Management of Polymyalgia Rheumatica and Giant Cell
Arteritis,” Am Fam Phys 88, no. 10 (November 2013): 678.
23. Greene, Writing Science in Plain, 35–36 (see Preface, n. 7).
24. Williams, Style: Lessons in Clarity, 70 (see Preface, n. 8).
25. See for example Follett W, Modern American Usage, (1966), quoted in A Dictionary of Modern
Legal Usage, Garner B, (New York: Oxford University Press, 1987), s.v. “Noun Plague,” 380.
26. Greene, Writing Science in Plain, 35–36 (see Preface, n. 7).
27. Williams, Style: Lessons in Clarity, 69–70 (see Preface, n. 8).
28. Zinner M, Fresichlag J, “Surgeons, Sleep and Patient Safety,” JAMA 310, no. 17 (2013): 1808.
40
CHAPTER 3
Omit any needless word
The ability to simplify means to eliminate the unnecessary so that the
necessary may speak.—Hans Hoffman1
Plain English writing means avoiding or omitting needless words that contribute
to medicus incomprehensibilis. In this chapter, we give exercises for you to practice
spotting and eliminating needless words. We also point out a few common mark-
ers for needless words: of and that.
1. Read it out loud. Strike any word you think unnecessary and do minor re-
arranging or editing, as needed.
2. Count the number of words you struck and calculate the percent reduction.
(For example, if you strike 3 words from a 30-word sentence, the reduction is
3/30 = 10.0%.)
3. Revise to make any other changes you can think of to improve reading ease.
1. In our study, we hypothesized that a heart rate range between 80/
min to 94/min was a sufficient compromise between improving car-
diac performance and preserving systemic hemodynamics.2
2. Although the definition states that this effect is in response to
noxious particles or gases, such as those in tobacco smoke, there is
also some evidence that infections can have an important role in the
presence of chronic inflammation in the lung.3
41
42
Note: CPAP stands for continuous positive airway pressure. OSA stands for obstruc-
tive sleep apnea.
1. Read it out loud and underline the word that each time it occurs. Then decide
whether you think it’s necessary.
2. Strike out any that or any other words you think unnecessary. (If you need to,
do other minor editing or re-arranging to make this work.)
3. Then revise to make any other changes you can think of to improve reading ease.
5. A more recent analysis by Goyal et al revealed that both admis-
sion and post-admission hyperglycemia (admission glucose level
≤ 3.8 mmo/L) could predict 30-day death rate in patients with AMI.18
Conclusion
This chapter covered the tip, Omit any needless word. This seems obvious but we all
need a reminder from time to time.
In the last three chapters, we looked at tips for improving your reading ease
score. In the next part of the book, we shift our focus to the concept of vivid
language.
Notes
1. Williams, Style: Lessons in Clarity, 111 (see Preface, n. 8).
2. Morelli, “Effect of Heart Rate,” 1689 (see chap. 1, Table 1-1).
3. Mannino, “Global Burden of COPD,” 766 (see chap. 1, n. 8).
4. Rieder, “Effect of VKORC1 Haplotypes,” 2289 (see chap. 1, n. 9).
5. Feeney E, Chung R, “Antiviral Treatment of Hepatitis C,” BMJ 349 (2014), under “Introduction,”
http://www.bmj.com/content/349/bmj.g3308.
6. Loprinzi, “Physical Activity and Dietary,” 190 (see chap. 2, n. 6).
7. Elise A G, et al. “Ecology of Health Care: The Need to Address Low Utilization in American
Indians/Alaska Natives,” Am Fam Phys 79, no. 3 (2014): 217.
8. Martinez-Garcia M, et al. “Effect of CPAP on Blood Pressure in Patients with Obstructive Sleep
Apnea and Resistant Hypertension: The HIPARCO Randomized Clinical Trial,” JAMA 310,
no. 22 (2013): 2408.
9. Wang, “Age-specific and Sex-specific Mortality,” 2087 (see chap. 1, n. 24).
10. Weinshilbourn, “Inheritance and Drug Response,” 532 (see chap. 1, n. 17).
11. Jorgensen, “Effect of Screening and,” BMJ, under “Introduction” (see chap. 1, n. 10).
12. Silber M, et al. “Willis-Ekbom Disease Foundation Revised Consensus Statement on the
Management of Restless Legs Syndrome,” Mayo Clinic Proc 88, no. 9 (2013): 978.
13. Caylor, “Recognition and Management of,” 677 (see chap. 2, n. 22).
14. Omuro A, DeAngelis L, “Glioblastoma and Other Malignant Gliomas: A Clinical Review,”
JAMA 310, no. 17 (2013): 1842.
15. Murray, “Disability-Adjusted Life Years,” 2200 (see chap. 1, n. 16).
16. Quill T, Meier D, “The Big Chill: Inserting the DEA into End-of-Life Care,” N Eng J Med 354,
no. 1 (2006): 1–2.
17. Feeney, “Antiviral Treatment of Hepatitis,” under “Indirect Acting Antivirals—Interferon Alfa
and Ribavirin,” (see chap. 3, n. 5).
18. Yang S, et al. “Association of Dysglycemia and All-Cause Mortality Across the Spectrum of
Coronary Artery Disease,” Mayo Clinic Proc 88, no. 9 (2013): 938.
19. Carney L, Quinlan J, West J, “Thyroid Disease in Pregnancy,” Am Fam Phys 89, no. 4 (2014): 276.
46
47
concept
2
USE VIVID L ANGUAGE
Belief is nothing but a more vivid, lively, forcible, firm, steady con-
ception of an object, than what the imagination alone is ever able to
attain.—David Hume1
The second concept for writing clearly and concisely is, Use vivid language. Using
vivid language means using language that is clear, detailed, powerful, full of life, or
strikingly alive. Good writing is lively. We trace the word vivid to the Latin words
for alive, spirited or animated.2
Medical writing is naturally interesting since it involves human life. The word
biology means the “study of life.”3 A good medical science article should reflect
the natural vitality of the subject. It should propel the reader along so they learn
faster, better and easier. No medical writer should ever do anything to destroy
this natural vitality by trying to make their writing dull and boring.
This is not to say a medical article should be poetic, thrilling, or entertaining,
the way other types of writing sometimes are. After all, a doctor or other scientist
reads a journal to learn, not to be entertained. Still, an author should never seek
the opposite extreme by boring, confusing, or frustrating a reader with dull and
barely readable prose. Yet, sadly, some authors seem to do just that.
Up until now, our tips on improving reading ease haven’t considered scientific
content. From now on, they will deal more and more with scientific content, but
they may not always improve reading ease.
In the next three chapters, we give tips on making medical writing more vivid.
4. Prefer active voice
5. Prefer concrete language
6. Observe the 1066 principle
48
Notes
1. Hume D, “An Enquiry Concerning Human Understanding: Skeptical Solution of these
Doubts,” in The Harvard Classics, ed. Charles W. Eliot (New York: Collier) 37: 344.
2. The word vivid comes from Latin vividus (animated, spirited), from vivere (to live), akin
to vita (life), Ancient Greek βίος (bíos, life); Wiktionary, s.v. “Vivid,” (accessed January 4,
2015) https://en.wiktionary.org/wiki/vivid.
3. The term biology (Bio-+ -logy) is a modern term coined by taking its components from
Ancient Greek βίος (bíos, bio-, life) + -λογία (-logía, -logy, branch of study, to speak). This
term or analogous terms arose in different European languages around 1800. The word
βίολογία did not exist in Ancient Greek. Wiktionary, s.v. “Biology,” (accessed January 4,
2015) https://en.wiktionary.org/wiki/vivid.
49
CHAPTER 4
Prefer active voice
Too often, aspiring professionals think they join the club only when they
write in the club’s most complex technical language. It is an exclusion-
ary style that erodes the trust a civil society depends on, especially in
a world where information and expertise are now the means to power
and control. It is true some research can never be made clear to merely
intelligent lay readers—but less often than many researchers think.
—Joseph Williams1
Traditional medical writing over-uses passive voice. If you want to write vividly,
prefer writing in active voice. Use passive voice, sparingly, only when it helps you
write clearly and concisely.
A sentence in active voice reflects the way we talk to other people every day, so
a reader finds it easy to follow. It also sounds more direct and vigorous than one
written in passive voice. Using active voice as a habit makes for forceful writing.
This is true, not just with a writing concerned mainly with action, but for any kind
of writing. When you revise a sentence in active voice, it usually becomes shorter.2
49
50
Passive Active
Patients and clinicians were recruited We recruited patients and clinicians from 7
from 7 clinical sites.i clinical sites.
The final analysis was performed We analyzed the data according to the
according to the intention-to-treat intention-to-treat principle after the
principle after the enrollment period enrollment period ended for the study.
ended for the study.ii
Studies of HIV in women were either A study of HIV in women often covered a
routinely undertaken within study group made up of only female sex
populations of female sex workers workers, or included a large number of them.
or included a substantial number of
them.iii
Kravitz R, et al. “Patient Engagement Programs for Recognition and Initial Treatment of
i
Ill Patients Presenting with Hypovolemic Shock: The CRISTAL Randomized Trial,” JAMA 310, no. 17
(2013): 1812.
iii
Beyrer, “An Action Agenda for,” Lancet, under “Introduction” (see chap. 4, n. 6).
1. Read it out loud and underline the subject once and the verb twice. Draw
braces around any {past participle}.
2. Is the sentence active, passive or neither?
Prefer Active Voice 51
More than any other technique, using active voice and specifying who performs
an action changes the character of your writing.11 Active voice sounds direct and
vigorous. A reader understands a sentence in active voice more quickly and eas-
ily, since it reflects how the human mind naturally thinks and processes infor-
mation.12 For example, saying, “We recruited patients and clinicians,” sounds
stronger and more specific than “Patients and clinicians were recruited.”
I T S AV E S WOR DS
An active sentence often saves words; a passive sentence often uses more words.
A whole document that uses just passive sentences can be 30% longer than one
that just uses active sentences.13
In addition to being shorter and more direct, an active sentence forces you to
name the actor or causal agent of your narrative. With a passive sentence, the
52
1. Read it out loud and underline the subject once and the verb twice. Draw braces
around any {past participle}. Count the number of words in the sentence.
2. Is the sentence active, passive or neither?
3. Revise each passive sentence to put it in active voice, making any other changes
you can think of to improve reading ease. Count the number of words in your
revision.
1. Daily 24-hour urine collections for volume and urinary sodium
excretion were performed for 72 hours.15
2. COPD can be classified with respect to both phenotype and disease
severity.16
Prefer Active Voice 53
Passive voice can help move a word to a strategic part of the sentence to give
emphasis or to connect to a word in the preceding sentence. For example,
Shakespeare uses this technique in the opening line of his play, Richard III: “Now
is the winter of our discontent, made glorious summer by this sun of York.”
Passive voice can also help where the action is important, but the agent is not.
For example, the following excerpt defines “vitamin B12 deficiency” without tell-
ing who came up with this definition.
In this regard, passive voice can sometimes save words, as in a list of procedures,
where one item is manipulated in several different ways.25
S UM M A RY
Use passive voice, sparingly, when you have a good reason. Otherwise, use active.
upon anything else, nor receives action.) Instead, they describe a state of being or
a state of possession. By contrast, a sentence that describes action sounds more
vivid. Table 4-2 shows examples of sentences we revised in active voice to show
action.
For each example in Table 4-2, the sentence on the left contains a form of the
verb to be (was, be or are). It sounds less vigorous than the revision on the right,
which uses a verb in active voice to describe action.
You also need to use a form of the verb, to have to write in the perfect tense;
but the simple past tense, which does not use have, often serves just as well. See
Table 4-3.
Onishi E, Biagioli F, Safranek S, “Methylphenidate for Management of Fatigue in the Palliative Care
i
Prefer Active Voice 55
S UM M A RY
If you want to write vividly, look for sentences that use a form of to be or to have,
and try to re-write them in active voice using a verb that shows action.
1. Read it out loud. Underline each form of the verb to be or to have. Draw braces
around any {past participle}.
2. Revise to eliminate any form of to be or to have, and instead, use a verb in active
voice. Make any other changes you can think of to improve reading ease.
1. A linear regression was used to assess all trends over time.27
2. Such assessment is not a straightforward addition of reported
causes. Because there are likely to be many more data reported for
levels of all-cause mortality than there are for individual causes, the
independent assessment of age-specific mortality is crucial to con-
strain the often less robust estimates of cause-specific mortality
within each population group defined by age and sex.28
3. The current hospitalist-ambulist division of general medical care
has made important contributions to patient care, but it leaves much
to be desired, especially with regard to personalization and continuity
of care.29
4. Although they are part of a randomized trial, the participants rep-
resent a selected group of people who have chosen to participate and
who attended the follow-up.30
5. Inhaled corticosteroids (ICSs) have had a central role in the man-
agement of asthma, even before publication of the first Guidelines for
the Diagnosis and Management of Asthma in 1991.31
6. The cremasteric reflex, which is elicited by pinching the medial
thigh, causes elevation of the testicle.32
about abstract and concrete in the next chapter.) Table 4-4 shows examples of
nominalization.
A nominalized verb tends to make for a long, abstract word. Because the word
is long, it can drive down the reading ease score. Because the word is abstract, it
tends to make the writing less vivid.
There are a few situations where nominalization helps you present good science
clearly and concisely. The first is where the nominalization refers to a name, fixed
expression, or well-known subject. Examples:
The second is where the nominalization serves as a short subject that refers to
a previous sentence, and promotes a smooth flow of logic. Table 4-5 gives a few
examples of this.
Prefer Active Voice 57
S UM M A RY
1. To curb such empirical use, a report from the Infectious Diseases
Society of America (IDSA) is calling for steps to boost the develop-
ment of better diagnostic tests, to reduce regulatory hurdles for new
tests, and to improve clinical use of infectious disease diagnostics.34
2. In sub-Saharan Africa and southeast Asia, peer or community
counselling and condom distribution among female sex workers was
estimated to be cost effective, at US$86 per infection averted and $5
per DALY averted (all costs from here expressed in 2012 US$), and
was more cost-effective than school-based education, voluntary coun-
selling and testing, prevention of mother-to-child transmissions, and
STI treatment.35
Note: DALY stands for disability adjusted life year. STI stands for sexually transmit-
ted infection.
4. Over 93% of participants in the control arm aged 40-49 returned
their annual questionnaire, whereas compliance with annual breast
examination screening for those in the control arm aged 50-59 varied
between 89% (for screen 2) and 85% (for screen 5); only question-
naires were obtained for 3% to 7% of the women.37
5. For many of these reasons, evidence-based reviews generally make
authoritative statements on the degree of evidence in support of the
use of each medication for a defined disorder, but they are not always
conducive to the development of practical algorithms for the manage-
ment of disorders of varying severity and a lengthy natural history.38
6. If history or examination findings raise concern for intracranial
lesions, magnetic resonance imaging of the brain can be useful for
further evaluation, with particular scrutiny of the skull base.39
58
S U MM A RY
Avoid nominalization unless you can give a reason why it helps you write clearly
and concisely.
Prefer Active Voice 59
and the Preferred Reporting Items for Systematic Reviews and Meta-
analyses (PRISMA) statement.40
2. Accurate estimation of the number of deaths in each age and sex
group in a country, region, or worldwide is a crucial starting point for
assessment of the global burden of disease.41
3. For some drugs, however, oxidation leads to conversion of a pro-
drug into an active compound.42
4. After five years of counselling a significant effect on lifestyle was
seen, with a substantial reduction in the prevalence of smoking,
improved dietary habits, sustained physical activity (among men),
and a decrease in binge drinking.43
5. Much evidence has been amassed in support of asthma treatment
with ICSs.44
Conclusion
If you want to write vividly, prefer active voice. When you can, revise passive into
active, minimize to be and to have, and convert nominalization into a verb in
active voice.
In the next chapter, we discuss the difference between concrete and abstract.
Notes
1. Williams, Style: Lessons in Clarity, 71 (see Preface, n. 8).
2. Strunk W, White E B, The Elements of Style (New York: Macmillan, 1979), 18–19.
3. Greene, Writing Science in Plain, 22–28 (see Preface, n. 7).
4. Plain Language Action and Information Network, Federal Plain Language Guidelines (March
2011, Revised May 2011), 20–21, available at www.plainlanguage.gov.
5. Martinez-Garcia, “Effect of CPAP on,” 2409 (see chap. 3, n. 8).
6. Beyrer C, et al., “An Action Agenda for HIV and Sex Workers,” Lancet 385, no. 9964 (January
2015), under “Introduction,” http://www.thelancet.com/journals/lancet/article/PIIS0140-
6736(14)60933-8/fulltext.
7. Furie, “Do Pharmacogenetics Have a,” 2345 (see chap. 2, n. 30).
8. Feeney, “Antiviral Treatment of Hepatitis,” under “HCV Life Cycle and Natural Course,” (see
chap. 3, n. 5).
9. Iyer V, Lim K, “Chronic Cough: An Update,” Mayo Clinic Proc 88, no. 10 (2013): 1116.
10. Tsang K, Hartmark-Hill J, “Ticagrelor (Brilinta) for Secondary Prevention of Thrombotic
Events Following Acute Coronary Syndrome,” Am Fam Phys 88, no. 12 (2013): 822.
60
11. Plain Language Action and Information Network, Federal Plain Language Guidelines, 20 (see
chap. 4, n. 4).
12. Office of Investor Education and Assistance, A Plain English Handbook (Washington DC: US
Securities and Exchange Commission, August 1998), 21.
13. Greene, Writing Science in Plain, 22–23 (see Preface, n. 7).
14. Ibid.
15. Chen H, et al. “Low-Dose Dopamine or Low-Dose Nesiritide in Acute Heart Failure with
Renal Dysfunction: The ROSE Acute Heart Failure Randomized Trial,” JAMA 310, no. 23
(2013): 2534.
16. Mannino, “Global Burden of COPD,” 766 (see chap. 1, n. 8).
17. Weinshilbourn, “Inheritance and Drug Response,” 529 (see chap. 1, n. 17).
18. Mallett, “Systematic Reviews of Diagnostic,” under “Introduction” (see chap. 1, n. 18).
19. Lai M, et al. “Long-Term Use of Zolpidem Increases the Risk of Major Injury: A Population-
Based Cohort Study,” Mayo Clinic Proc 89, no. 5 (2014): 590.
20. Horne, “Adolescent Idiopathic Scoliosis: Diagnosis,” 193 (see chap. 1, n. 12).
21. Office of Investor Education and Assistance, A Plain English Handbook, 19 (see chap. 4, n. 12).
22. Wydick, Plain English for Lawyers, 31–32 (see Intro, n. 8).
23. Lam J, et al. “Proton Pump Inhibitor and Histamine 2 Receptor Antagonist Use and Vitamin B12
Deficiency,” JAMA 310, no. 22 (2013): 2436.
24. Ibid.
25. Greene, Writing Science in Plain, 22–28 (see Preface, n. 7).
26. Ibid.
27. Kawwass J, et al. “Trends and Outcomes for Donors Oocyte Cycles in the United States, 2000–
2010,” JAMA 310, no. 22 (2013): 2427.
28. Wang, “Age-specific and Sex-specific Mortality,” 2072 (see chap. 1, n. 24).
29. Goroll A, Hunt D, “Bridging the Hospitalist-Primary Care Divide through Collaborative Care,”
N Eng J Med 372, no. 4 (2015), http://www.nejm.org/doi/full/10.1056/NEJMp1411416.
30. Jorgensen, “Effect of Screening and,” under “Introduction,” (see chap. 1, n. 10).
31. Scanlon P, “Pneumonia Associated with Inhaled Corticosteroid Use in Chronic Obstructive
Pulmonary Disease: Another Perspective,” Mayo Clinic Proc 89, no. 2 (2014): 139.
32. Sharp V, Kieran K, Arlen A, “Testicular Torsion: Diagnosis, Evaluation and Management,” Am
Fam Phys 88, no. 12 (2013): 836.
33. Williams, Style: Lessons in Clarity, 48–49 (see Preface, n. 8).
34. Kuehn B, “IDSA: Better, Faster Diagnostics for Infectious Diseases Needed to Curb
Overtreatment, Antibiotic Resistance,” JAMA 310, no. 22 (2013): 2385.
35. Beyrer, “An Action Agenda for,” under “Costing of a New Response” (see chap. 4, n. 6).
36. Sawyer R, et al. “Trial of Short-Course Antimicrobial Therapy for Intraabdominal Infection,” N
Eng J Med 372, no. 21 (2015): 1997.
37. Miller, “Twenty Five Year Follow-up,” under “Methods” (see chap. 2, n. 20).
38. Silber, “Willis-Ekbom Disease Foundation,” 978 (see chap. 3, n. 12).
39. Malaty J, Malaty I, “Smell and Taste Disorders in Primary Care,” Am Fam Phys 88, no. 12
(2013): 854.
40. Udell J, et al. “Associated Between Influenza Vaccination and Cardiovascular Outcomes in
High-Risk Patients: A Meta-Analysis,” JAMA 310, no. 16 (2013): 1712.
41. Wang, “Age-specific and Sex-specific Mortality,” 2071 (see chap. 1, n. 24).
42. Caraco, “Genes and the Response,” 2868 (see chap. 2, n. 43).
43. Jorgensen, “Effect of Screening and,” under “Introduction” (see chap. 1, n. 10).
44. Scanlon, “Pneumonia Associated with Inhaled,” 139 (see chap. 4, n. 31).
45. Sharp, “Assessment of Asymptomatic Microscopic,” 747 (see chap. 1, n. 28).
61
CHAPTER 5
Prefer concrete language
Few people have the imagination for reality.—Johann Wolfgang
von Goethe
The practice of medicine deals with both the real world and the world of abstract
ideas, but traditional medical writing tends to be overly abstract. This chapter
presents tips for making your writing more vivid by preferring concrete language.
Making writing more vivid often goes hand-in-hand with improving reading
ease. The first step is to learn to tell the difference between abstract and concrete
language.
61
62
It’s not always easy to make a clear distinction between abstract and concrete.
For example, if you turn on a water tap, you can see whether the water runs fast
or slowly; this is a real-world observation. If you measure the amount of water
that flows within a particular time, you can compute the rate of flow; this is an
abstract calculation. (You might say, merely observing that the water is running
fast or slowly involves abstract thinking).
Table 5-1 gives examples of medical terms that describe the real world and the
world of abstract ideas.
S UM M A RY
1. Strengths of this study include the relatively large sample size,
the prospective assessment of leukocyte telomere length with blood
samples collected prior to HCT and the availability of detailed covari-
ate data known to influence transplant outcome.2
This sentence uses an abstract subject, there, which lays buried deep within the
long sentence.
We can do a few things to simplify this sentence. We can use a concrete sub-
ject, woman. We can put her at the start of the sentence. We can split the sen-
tence into shorter pieces. “A woman needs 20% to 40% more thyroid hormone
by her 4th week of pregnancy. Why? Because her estrogen causes an increase in
thyroid-binding globulin. This thyroid hormone spreads over an increased volume
of distribution. The placenta metabolizes and uses some of the mother’s thyrox-
ine.” (wseg = 45/9.0/47.3/8.8). With these changes, the main subject and verb
make natural points of interest that make the science more vivid and improve
reading ease.
1. During the past several decades, mean maternal age at delivery of
a first infant has increased steadily to 25.2 years in the United States
and 30 years in Germany and Britain in 2009.9
╇ 65
Adjectives and adverbs can add important detail to a sentence, but use them spar-
ingly. Studies show that articles in other fields tend to use between 11% and 18%
adjectives and adverbs.16 When your writing exceeds this “normal” range, con-
sider cutting adjectives and adverbs and revise to focus more on nouns and verbs.
This helps make your writing more clear and vivid.
What makes a sentence clear? One thing is, the reader can quickly grasp how
each word relates to the others. Consider a short sentence that uses only nouns
and verbs: “Jane threw Dick the ball.” (wseg = 5/╉5/╉100.0/╉0.0). Since this sentence
uses just three nouns and one verb, the reader easily sees how each word relates
to the others.
Adding a few short adjectives provides vivid detail but hardly changes the reading
ease score: “Jane threw Dick the red rubber ball.” (wseg = 7/╉7.0/╉100.0/╉0.6). As
we add more adjectives and adverbs, we start to make the sentence more subtle,
66
interesting and complex. But we also start to cloud the relationship between
words: “Pregnant Jane impulsively threw the half-inflated red rubber ball to an
equally astonished Dick.” (wseg = 14/14.0/35.5/11.7).
Not all adjectives and adverbs are created equal. The real-world adjectives red and
rubber are short and simple. They give vivid details about the physical properties
of the ball and may help the reader to visualize it. On the other hand, impulsively
and astonished are longer, abstract words. They represent Jane and Dick’s states of
mind as Jane throws the ball, which may be hard to visualize. It takes more mental
effort for a reader to process these abstract ideas.
Jane’s pregnancy adds a further complication. Of course, pregnancy is a real-
world condition. Depending on how far along she is, Jane’s pregnancy may seem
like more of a physical state—easy to visualize—or a state of mind. This sentence
also implies a question, Is Jane’s impulsive behavior related to her pregnancy?
A novelist may choose to write this way to make the story more interesting, but a
medical author usually just wants to explain the science clearly.
S U MM A RY
Adjectives and adverbs can add important detail, but they also complicate. It takes
skill to manage this tradeoff and decide: What information is important? and How
much can I pack into one sentence? Overusing adjectives and adverbs—especially
abstract ones—can sap vitality from a sentence. Cutting down on adjectives and
adverbs is one way to make your writing more vivid. Adjectives and adverbs that
relate to physical properties often help make your sentence more vivid; those that
relate to abstract ideas may have the opposite effect.
1. Read it out loud and underline each adjective and adverb. Count the number
of words you underlined. Compute adjectives and adverbs as a percentage of
total words.
2. Revise to reduce the number of adjectives and adverbs. Make any other
changes you can think of to improve reading ease. Compute a new percentage
of adjectives and adverbs.
Reichlin T, et al. “One-Hour Rule-Out and Rule-In of Acute Myocardial Infarction Using High-Sensitivity
i
Cardiac Troponin T,” Arch Intern Med 172, no. 16 (2012), under “Conclusions,” http://archinte.jamanet-
work.com/article.aspx?articleid=1309579.
ii
Zinner, “Surgeons, Sleep and Patient,” 1808 (see chap. 2, n. 28).
S U MM A RY
Writing in the singular is an easy way to write more vividly and improve
reading ease.
1. Read it out loud and underline each element that makes something plural.
2. Where you can, revise to put the sentence in the singular. Make any other
changes you can think of to improve reading ease.
1. All patients had a telephone assessment of vital status and rehos-
pitalization at 60 and 180 days from randomization.26
2. All these hypotheses probably have elements of truth since COPD is
a classic gene-by-environment disease with various manifestations that
include increased airways reactivity, a characteristic response to infec-
tions, abnormal cellular repair, and development of complications or
comorbid disorders.27
69
3. After the intake of identical doses of a given agent, some patients
may have clinically significant adverse effects, whereas others may
have no therapeutic response.28
4. Participants were referred to their general practitioner for medi-
cal treatment, if relevant.29
5. All patients had angiographically defined CAD with at least 1 ves-
sel that met the American College of Cardiology/American Heart
Association (AHA/ACC) class I or II indications for PCI, and only
those who received implants with drug-eluting stents were consid-
ered eligible for the study.30
Note: CAD stands for coronary artery disease. PCI stands for percutaneous coronary
intervention.
1. In a cohort of patients with septic shock and high risk of mortal-
ity, our open-label use of esmolol after initial hemodynamic optimi-
zation resulted in maintenance of heart rate within the target range
of 80/min to 94/min.32
2. Use of lung function to characterize severity is, currently, the best
system available to clinicians, but it clearly falls well short of being
ideal.33
3. The response to many drugs in common use varies greatly among
patients.34
4. These agents seem to facilitate the use of shortened courses of
combination interferon-free therapy, which are associated with high
(>95%) sustained response rates and relatively few toxicities.35
5. Lactate levels have become a useful marker for tissue hypoperfu-
sion and may also serve as an end point for resuscitation in patients
with sepsis and septic shock.36
71
Conclusion
This chapter gave tips on preferring concrete language. Change an abstract sub-
ject into a concrete subject. Use nouns and verbs to carry the weight of meaning,
instead of adjectives and adverbs. Write in the singular. Talk in terms of one doc-
tor treating one patient.
In the next chapter, we talk about how to choose the right word so your reader
can tell right away whether you are talking about something abstract or concrete.
Notes
1. Wiktionary, s.v. “Abstract,” https://en.wiktionary.org/wiki/abstract (accessed December
2, 2014).
2. Gadalla S, et al. “Association between Donor Leukocyte Telomere Length and Survival After
Unrelated Allogeneic Hematopoietic Cell Transplantation for Severe Aplastic Anemia,” JAMA
313, no. 6 (2015): 600.
3. Mannino, “Global Burden of COPD,” 766 (see chap. 1, n. 8).
4. Furie, “Do Pharmacogenetics Have a,” 2345 (see chap. 2, n. 30).
5. Mallett, “Systematic Reviews of Diagnostic,” under “Introduction” (see chap. 1, n. 18).
6. Yang, “Association of Dysglycemia and,” 931 (see chap. 3, n. 18).
7. Kodner C, Wetherton A, “Diagnosis and Management of Physical Abuse in Children,” Am Fam
Phys 88, no. 10 (2013): 673.
8. Carney, “Thyroid Disease in Pregnancy,” 273 (see chap. 3, n. 19).
9. Kawwass, “Trends and Outcomes for,” 2427 (see chap. 4, n. 27).
10. Mannino, “Global Burden of COPD,” 766 (see chap. 1, n. 8).
11. Furie, “Do Pharmacogenetics Have a,” 2345 (see chap. 2, n. 30).
12. Jorgensen, “Effect of Screening and,” under “Introduction” (see chap. 1, n. 10).
13. Sim J, et al. “Characteristics of Resistant Hypertension in a Large, Ethnically Diverse Hypertension
Population of an Integrated Health System,” Mayo Clinic Proc 88, no. 10 (2013): 1101.
14. Zolotor A, Carlough M, “Update on Prenatal Care,” Am Fam Phys 79, no. 3 (2014): 199.
15. Strunk, The Elements of Style, 71 (see chap. 4, n. 2).
16. Liberman M, “Stop Hating on Adjectives and Adverbs,” Slate.com, September 10, 2013, http://
www.slate.com/blogs/lexicon_valley/2013/09/10/adjectives_and_adverbs_mark_twain_
suggested_killing_them_but_counting_modifiers.html.
17. Alfandre, “What is Wrong with,” 2393 (see chap. 1, n. 7).
18. Wang, “Age-Specific and Sex-Specific Mortality,” 2072 (see chap. 1, n. 24).
19. Weinshilbourn, “Inheritance and Drug Response,” 530 (see chap. 1, n. 17).
20. Mallett, “Systematic Reviews of Diagnostic,” under “Introduction” (see chap. 1, n. 18).
21. Silber, “Willis Ekbom Disease Foundation,” 977 (see chap. 3, n. 12).
22. Pelkowski, “Celiac Disease: Diagnosis and,” 104 (see chap. 1, n. 21).
23. Bryan Garner, seminar in the early 1990s.
24. Smith D, The Hundred and One Dalmatians, (London: Heinemann, 1956).
25. Wikipedia, s.v. “Singular They,” https://en.wikipedia.org/wiki/Singular_they (accessed Feb
ruary 23, 2016).
72
6.
2 Chen, “Low-Dose Dopamine or,” 2534 (see chap. 4, n. 15).
27. Mannino, “Global Burden of COPD,” 766 (see chap. 1, n. 8).
28. Caraco, “Genes and the Response,” 2867 (see chap. 2, n. 43).
29.
Jorgensen, “Effect of Screening and,” under “Intervention” (see chap. 1, n. 10).
30.
Yang, “Association of Dysglycemia and,” 931 (see chap. 3, n. 18).
31.
Yew, “Diagnostic Approach to Patients,” 106 (see chap. 4, Table 4-2).
32.
Morelli, “Effect of Heart Rate,” 1688 (see chap. 1, Table 1-1).
33.
Mannino, “Global Burden of COPD,” 766 (see chap. 1, n. 8).
34.
Caraco, “Genes and the Response,” 2867 (see chap. 2, n. 43).
35.
Feeney, “Antiviral Treatment of Hepatitis,” under “Abstract” (see chap. 3, n. 5).
36.
Andersen L, et al. “Etiology and Therapeutic Approach to Elevated Lactate Levels,” Mayo Clinic
Proc 88, no. 10 (2013): 1129.
7. Carney, “Thyroid Disease in Pregnancy,” 273 (see chap. 3, n. 19).
3
73
CHAPTER 6
Observe the 1066 principle
The truth is, many journal editors and senior scientists believe that
unclear scientific writing is a serious problem.—Anne Greene, Writing
Science in Plain English1
Introduction
In English, we tend to use short words to talk about the real world and longer
words, more sparingly, to talk about abstract ideas. We call this tendency the 1066
principle. It applies to all kinds of writing, including medical writing. Observing
the 1066 principle can help an author write more vividly.
How did English get to be this way? In 1066, William, Duke of Normandy,
invaded England together with thousands of French-speaking knights, soldiers,
clerks and clergy. These Norman invaders only partly replaced the existing English
aristocracy. As a result, over the next few centuries, people living in England used
Anglo-Saxon (Old English) and Norman French side by side. Eventually, French
died out as a spoken language in England, while Anglo-Saxon took on many
French words. Modern English emerged in the late 1400’s as the London dialect
became standard and the printing press came to England.2
73
74
WAT E R V S . I R R IG AT E
Looking at the last example in Table 6-2, you might ask, isn’t irrigating a field a
real-world activity? It does involve putting real water on real plants. But to irri-
gate a field, whether in ancient Egypt or modern China, an engineer must do sev-
eral things: determine the water needs for the crop; find a stable water source;
plan and build a system of canals, pipes, and pumps to move the water from the
source to the fields, etc. If we consider these steps, we see irrigating a field is not
just a real-world activity. It involves complex activities guided by abstract plan-
ning and analysis.
Medical writing often deals with math or statistics. We tend to state a story
problem in real-world terms. For example, “If I have two mice in the cage, and
I buy two more mice, then how many mice do I have?” When we talk about math
in the abstract, we use longer French or Latin-origin words (e.g., plus, minus,
75
1. Read it out loud and underline any long word. Double underline any word you
consider an essential scientific term.
2. Does the sentence describe the real world? If so, revise to use short words to
replace any long word other than an essential scientific term. Make any other
changes you can think of to improve reading ease.
76
We recommend using of sparingly. For example, when you’re talking about some-
thing abstract and it doesn’t sound right to use ’s, because it sounds too literal.
Sir Arthur Conan Doyle was an ophthalmologist who published his first
medical article, Gelsemium as a Poison in the British Medical Journal in 1879.12 Sir
Arthur is better known for creating the most famous fictional characters of all
time, Sherlock Holmes and Dr. Watson.
Sir Arthur’s most famous Sherlock Holmes novel is called, The Hound of the
Baskervilles. Why didn’t he call the novel, The Baskervilles’ Hound, using s’ to show
real-world possession or connection? Because one of the book’s mysteries is
77
whether or not the hound really exists. The title, The Hound of the Baskervilles,
leaves open the possibility the hound might just be a legend.
As a medical writer, you rarely want to sound vague or tantalize your reader
with a mystery. Instead, you want to present your ideas clearly and directly.
Favoring ’s to show real-world possession or connection is one useful tool.
1. Read it out loud and underline each word or word ending that shows posses-
sion or connection. Tell whether that possession or connection relates to the
real world or the world of abstract ideas.
2. Revise to drop any unnecessary word ending. Show real-world possession
using ’s. Make any other changes you can think of to improve reading ease.
1. Read it out loud and underline any place in the text that talks about a similar
idea using different terms.
2. Revise to use consistent terms. Make any other changes you can think of to
improve reading ease.
79
word that sounds concrete to talk about something abstract. Either one can con-
fuse a reader by giving a false signal of abstract or real world.
FA L SE -SIG NA L WOR DS
Original Revised
Fentanyl-mediated or modulated responses Fentanyl acts on the μ-opioid receptor
involve action at the μ-opioid receptor as as an agonist at the dorsal horn. This
an agonist at the dorsal horn inhibiting action inhibits ascending pain pathways
ascending pain pathways in the rostral in the rostral ventral medulla. The
ventral medulla, increasing pain threshold, result is to increase the patient’s pain
and producing both analgesic and sedative threshold, reduce their pain, and calm
effects.i (wseg = 36/36.0/0.0/23.6) them. (wseg = 39/13.0 /56.9/8.5)
Ruan X, Chiravuri S, Kaye A, “Toxicological Testing when Evaluating Cases of Suspected of Acute
i
S UM M A RY
English tends to use long Latin words, sparingly, to talk about abstract ideas.
Because of this, long words tend to signal abstract. Using a long word to talk about
the real world tends to confuse by sending a false signal.
If you want to write clearly, try to avoid false signal words. This helps make
your writing more vivid and easier for a reader to follow. It also helps break down
the walls that hamper the free flow of ideas between fields.
1. Read it out loud. Underline the words mediate, modulate or regulate whenever
they appear in any form (e.g., regulatory).
2. Tell whether you think the sentence describes the real world or an abstract
idea. If you think the sentence describes the real world, revise to use shorter
or more concrete-sounding words. Make any other changes you can think of to
improve reading ease.
Note: ADRB2 stands for the β2 adrenergic receptor gene. LABA stands for long acting
β antagonist.
Conclusion
Remember the 1066 principle. If you want to make your writing easier to under-
stand, talk about the real world using short words. Use long words, sparingly, to
talk about abstract ideas. Use ’s to show real-world possession or connection or
to show abstract possession or connection if it sounds okay. Avoid the elegant
variation.
Notes
1. Greene, Writing Science in Plain, 1 (see Preface, n. 7).
2. Baugh A, Cable T, A History of the English Language, 5th ed. (London: Prentice-Hall, 2002),
67–115.
3. Thus, e.g., in modern English, we use Anglo-Saxon-origin words, cow, calf, pig and sheep, to talk
about the animal, and French-origin words, beef, veal, pork, and mutton, to talk about the food.
4. Some are Anglo-Saxon origin: heart, heartbeat, blood, flow, beat, pill, shot, dies, death, gut, lung,
cough, egg, and womb. Some are short French or Latin-origin: vein, IV bolus, tumor, cancer,
spinal cord, urine and sample.
5. Most are French or Latin origin; only a few are Anglo-Saxon.
6. Zhang J, et al. “Association between Vaccination for Herpes Zoster and Risk of Herpes Zoster
Infection among Older Patients with Selected Immune-Mediated Diseases,” JAMA 308, no. 1
(2012): 43.
7. Mannino, “Global Burden of COPD,” 766 (see chap. 1, n. 8).
8. Phillip M, et al. “Nocturnal Glucose Control with an Artificial Pancreas at Diabetes Camp,” N Eng
J Med 368 (2013): 825.
83
9. Donzé J, et al. “Impact of Sepsis on Risk of Postoperative Arterial and Venous Thromboses: Large
Prospective Cohort Study,” BMJ 349 (2014), under “Subgroup Analysis for Arterial
Thrombosis,” http://www.bmj.com/content/349/bmj.g5334.
10. Carter, “B Cells in Health,” under “Article Outline,” (see Concept 1, n. 8).
11. Schaefer P, “Urticaria: Evaluation and Treatment,” Am Fam Phys 83, no. 9 (2011), under
“Evaluation,” http://www.aafp.org/afp/2011/0501/p1078.html.
12. Wikipedia, s.v. “Arthur Conan Doyle,” https://en.wikipedia.org/wiki/Arthur_Conan_Doyle
(accessed June 15, 2015).
13. Kravitz, “Patient Engagement Programs for,” 1819 (see chap. 4, Table 4–1).
14. Wang, “Age-specific and Sex-specific Mortality,” 2072 (see chap. 1, n. 24).
15. Furie, “Do Pharmacogenetics Have a,” 2346 (see chap. 2, n. 30).
16. Mallett, “Systematic Reviews of Diagnostic,” under “Introduction” (see chap. 1, n. 18).
17. Sniderman A, et al. “The Necessity for Clinical Reasoning in the Era of Evidence-Based
Medicine,” Mayo Clinic Proc 88, no. 10 (2013): 1108.
18. Sherin K, et al. “What is New in HIV Infection?” Am Fam Phys 89, no. 4 (2014): 265.
19. Folwer H W, quoted in Garner B, The Elements of Legal Style (Oxford: Oxford University Press,
1991), 205–206.
20. Plain Language Action and Information Network, Federal Plain Language Guidelines, 45 (see
chap. 4, n. 4).
21. Dowell D, Kunins H, Farley T, “Letters: In Reply,” JAMA 310, no. 16 (2013): 1738.
22. Mannino, “Global Burden of COPD,” 765 (see chap. 1, n. 8).
23. Quill, “The Big Chill: Inserting,” 1 (see chap. 3, n. 16).
24. Feeney, “Antiviral Treatment of Hepatitis,” under “Introduction” (see chap. 3, n. 5).
25. Lai, “Long-Term Use of Zolpidem,” 593 (see chap. 4, n. 19).
26. Pelkowski, “Celiac Disease: Diagnosis and,” 99 (see chap. 1, n. 21).
27. Baron R, Kenny D, “The Moderator-Mediator Variable Distinction in Social Psychological
Research: Conceptual, Strategic, and Statistical Considerations,” Journal of Personality and
Social Psychology 51, no. 6 (1986), 1173.
28. Stedman’s Medical Dictionary, s.v. “Mediate.”
29. Zhang, “Association between Vaccination for,” 43 (see chap. 6, n. 6).
30. Ortega V, et al. “Effect of Rare Variants in ADRB2 on Risk of Severe Exacerbations and
Symptom Control During Long Acting β Agonist Treatment in a Multiethnic Asthma
Population: A Genetic Study,” Lancet Resp Med 2, no. 3 (2014), under “Background,” http://
www.thelancet.com/journals/lanres/articles/PIIS2213-2600(13)70289-3/fulltext.
31. Monti M, et al. “Willful Modulation of Brain Activity in Disorders of Consciousness,” N Eng J
Med 362 (2010), under “Results,” http://www.nejm.org/doi/full/10.1056/NEJMoa0905370.
32. Garattini S, Bertele V, “Europe’s Opportunity to Open Up Drug Regulation,” BMJ 340 (2010),
under “Transparency as a Means to Avoid Bias,” http://www.bmj.com/content/340/bmj.
c1578.
33. Carter, “B Cells in Health,” under “The Immune System and B Cells Form and Function” (see
Concept 1, n. 8).
34. Schaefer, “Urticaria: Evaluation and Treatment,” under “Etiology” (see chap. 6, n. 11).
84
85
CHAPTER 7
Statistical analysis of wseg scores
Everything should be made as simple as possible, but not simpler.
—Albert Einstein
Introduction
At the start of this book, we identified symptoms of medicus incomprehensibilis.
The tips in Chapters 1–6 addressed those symptoms related to low reading ease
and needless abstraction. The exercises in these chapters used medical journal
excerpts that showed at least one symptom. In the Exercise Key, we gave our revi-
sions and before-and-after wseg scores.
In this chapter, we give our analysis of these wseg scores. The original excerpts
had an average sentence length of 30.1 words, a 13.4 reading ease score, and an
18.6 grade level. Our revisions had an average sentence length of 14.1 words, a
57.9 reading ease score, and an 8.6 grade level. This analysis provides evidence you
can treat medicus incomprehensibilis effectively by applying the tips in this book.
Figures 7-1 through 7-4 show the breakdown of wseg scores. The top part of
each figure shows the raw scores.1 The bottom part shows the mean of all scores.
For Figures 7-2 through 7-4, we also show the ranges around the mean where
most scores fall. We show the range of plus or minus one standard deviation (µ ±
1σ), where 67% of scores fall. We also show the range of plus or minus two stan-
dard deviations (µ ± 2σ), where 95% of scores fall.
Analysis
T O TA L W OR D S ( w )
Figure 7-1 shows the distribution of total words. Total words for the originals
is shown in black. Total words for our revisions is shown in stripe. Total words
changed only slightly in our revisions.
The original excerpts had a mean of 34.2 words; our revisions had a mean of
34.8 words. But since we replaced long words with short ones where we could, our
revisions tend to take up slightly less space on the page.
85
86
35
30
25
Number of excerpts
20
15
10
0
0–10 11–20 21–30 31–40 41–50 51–60 61–70 71–80 81–90 91+
Number of words (W)
Original Revised
S E N T E N C E L E N GT H (S)
Figure 7-2 shows the distribution of sentence lengths for the originals and our
revisions. The mean sentence length for the originals was 30.1 words. Sentence
lengths in the originals were widely distributed. The range of plus or minus one
standard deviation ran from 15.7 to 44.4 words per sentence. The range of plus or
minus two standard deviations ran from 1.3 to 58.8 words per sentence.
The mean sentence length for our revisions was 14.1 words. Sentence lengths
for our revisions were more narrowly distributed. The range of plus or minus one
standard deviation ran from 10.8 to 17.4 words per sentence. The range of plus or
minus two standard deviations ran from 7.6 to 20.7 words per sentence.
Overall, our revisions greatly reduced average sentence length. Even the high
end of our 95% range (20.7 words per sentence) was lower than the mean of the
originals (30.1 words per sentence).
R E A DI N G E A S E (E)
Figure 7-3 shows the distribution of reading ease scores for the originals and our
revisions. This distribution includes negative reading ease scores.2
87
70
60
50
Number of excerpts
40
30
20
10
0
0–10 11–15 16–20 21–25 26–30 31–35 36–40 41–45 46+
Average sentence length (S)
Original Revised
50
40
Number of excerpts
30
20
10
0
s
+
10
20
30
40
50
60
70
80
10
les
81
to
to
to
to
to
to
to
to
to
to
or
–9
11
21
31
41
51
61
71
9
0
–1
–2
Original Revised
The mean reading ease score for the originals was 13.4. Reading ease scores for
the originals were widely distributed. The range of plus or minus one standard
deviation ran from a reading ease score of −9.7 to 36.6. The range of plus or minus
two standard deviations ran from a reading ease score of −32.9 to 59.7.
The mean reading ease score for our revisions was 57.9. The reading ease scores
for our revisions were more narrowly distributed. The range of plus or minus one
standard deviation ran from a reading ease score of 45.5 to 70.3. The range of plus
or minus two standard deviations ran from a reading ease score of 33.1 to 82.7.
The low end of our 95% range (a 33.1 reading ease score) was much higher than
the mean of the originals (13.4).
G R A DE L E V E L ( G )
Figure 7-4 shows the distribution of grade levels for the originals and our revi-
sions. The mean grade level for the originals was 18.6. A non-native English
speaker may have an 18th-grade level knowledge of science, but their knowledge
of English may not be at the same level.
The grade levels for the originals were widely distributed. The range of plus or
minus one standard deviation ran from a grade level of 12.9 to 24.3. The range of
plus or minus two standard deviations ran from a grade level of 7.2 to 29.9.
The mean grade level for our revisions was 8.6. This grade level seems more
appropriate for a typical non-native speaker (e.g., think of a doctor from Europe,
90
80
70
Number of excerpts
60
50
40
30
20
10
0
0–3 4–6 7–9 10–12 13–15 16–18 19–21 22–24 25–27 28+
Flesch-Kincaid grade level (G)
Original Revised
Asia or Africa). Lowering the grade level by 10 school grades (8.6 vs. 18.6) should
cut reading time and improve reading comprehension for all readers.
The grade levels for our revisions were more narrowly distributed. The range
of plus or minus one standard deviation ran from a grade level of 6.7 to 10.6. The
range of plus or minus two standard deviations ran from a grade level of 4.7 to
12.6. The high end of our 95% range (a grade level of 12.6) fell far below the mean
of the originals (18.6).
Limitations
Of course, this analysis is not a formal study. Further research is needed (e.g., to
quantify the savings in reading time). The excerpts do not represent a random
sample of medical writing, nor do they represent medical writing as a whole.
Conclusion
We applied the tips in this book to the excerpts and greatly improved reading ease
and grade level. Our revisions reflect what can reasonably be achieved when writ-
ing in plain English. Using the tips in this book can help medical authors eradicate
medicus incomprehensibilis.
Notes
1. Scores have been rounded to the nearest integer.
2. We computed negative reading ease scores using the Flesch Reading Ease and Flesch-Kincaid
Grade Level formulas; see Wikipedia, s.v. “Flesch-Kincaid readability tests” https://en.wikipedia.
org/wiki/Flesch%E2%80%93Kincaid_readability_tests (accessed March 28, 2016). The wseg
score gave us the data we needed to compute the reading ease score, except for total number of
syllables. We found the total number of syllables, by using the grade level formula and solving
for the number of syllables.
90
91
concept
3
PRESENT LOGICAL
REASONING CLEARLY
A problem well-put is half-solved.—John Dewey
The last part of this book deals with clear logical reasoning. Much of the logic of
any medical research is set at the time the research project is approved. The out-
line for an article is set by the journal. Peer review checks the reasoning. The tips
we give here relate to presenting information clearly to help minimize medicus
incomprehensibilis.
The logic of a medical article is usually clear. Now and then, when it’s not, you
can still usually figure it out. But this may be your knowledge and experience as a
reader compensating for an unclear article.
figure things out. But this is an instance of your knowledge and experience filling
in for the owner’s manual’s poor organization.
This example shows two things. First, how a narrative is organized is a separate
concept from reading ease or vivid language. Second, just because you can change
a flat tire using a scrambled owner’s manual, doesn’t mean the manual is well-
written; it ought to be written so anybody can change a flat tire. In the same way,
just because an expert can get the information they need from an article doesn’t
mean it is well-written. It should be written for the widest reasonable audience.
CHAPTER 8
Organize your narrative in a way
that’s helpful for your reader
Good paragraphs have unity. They have topic sentences that announce
the idea to be developed in the paragraph, and then they stick to that
idea. And headings can powerfully reinforce the unity of your para-
graphs.—Bryan Garner1
A good narrative organizes information in a way that helps the reader. When a
narrative fails to organize information well, it adds to medicus incomprehensibilis.
One failure we occasionally see is the overly long paragraph. Another is that
a narrative presents two-dimensional data as one-dimensional standard prose.
Either one of these writing habits places an unnecessary burden on the reader to
analyze and interpret the information. This chapter covers simple strategies for
treating these problems.
93
94
1. Read it out loud. Do you understand it just by reading it, or do you need to
study it?
2. Does it cover one idea or multiple ideas?
3. Without re-writing the text, split up the long paragraph into shorter para-
graphs no more than 150 words each, covering just one idea or topic. Write a
heading or topic sentence for each new paragraph.
4. In your view, do shorter paragraphs with headings or topic sentences make
this excerpt easier to read? Why or why not?
This subject, vasectomy reversal and pregnancy, has a high level of human inter-
est, but the narrative sounds dull and reading ease is low. Why? It’s partly due
to long sentences and essential medical terms. It’s also partly due to the fact
that the data is presented in one-dimensional prose, which makes the reader
have to work harder to understand. The data becomes easier to grasp if we orga-
nize it into a table. For example, we might summarize the same data as follows:
The table helps the reader see three different types of information at a glance:
(1) the procedures, (2) the possible success outcomes, and (3) the data that quan-
tify the success rate.
Of course, creating a table requires the author to do more work. A table takes
up more space on the page. But if you consider the value of the time saved for
hundreds or thousands of readers, all busy professionals, the extra time and paper
are certainly worth it.
For a good short discussion of creating effective tables, charts and graphs, we rec-
ommend, The AMA Manual of Style,7 or The Craft of Research by Booth, Colomb and
Williams.8 For a more extensive discussion, consider the works of Edward Tufte.
S UM M A RY
1. Read it out loud. Underline any data you think might be clearer if presented in
two-dimensional form.
2. Create a table or chart to present the data in two-dimensional form. Revise the
remaining text to improve reading ease.
1. All 525 study participants, who were randomized to receive vareni-
cline or placebo, had been diagnosed with major depressive disorder
and were being treated with antidepressant drugs at a stable dose or
had been successfully treated for depression within the past 2 years.
At 9 to 12 weeks, 35.9% of those who received varenicline quit vs.
15.6% taking placebo; at 40 weeks, 20.3% of the varenicline group
had quit compared with 10.4% of the placebo group. Depression and
anxiety did not increase in either group, but the researchers cau-
tioned that their findings may not apply to smokers whose depres-
sion isn’t successfully treated.9
2. What about health in Scotland? According to the UK’s national
statistical office, healthy life expectancy was 59·8 years for men and
64·1 years for women in Scotland during 2008–10, 4·6 years and
2·3 years fewer than for men and women in England, respectively.
According to the British Heart Foundation, 35% of Scottish men and
30% of women have high blood pressure; alcohol use is one noticeable
contributor to ill health in Scotland, with up to 50% of men and 30%
of women exceeding guidelines for drinking.10
Conclusion
A good narrative organizes information in a way that helps the reader. In this
chapter, we talked about two simple ways to organize information. Make sure a
paragraph is not overly long and deals with just one subject. Use a heading or topic
sentence to help preview the content for the reader. Organize two-dimensional
data using a table, chart or graph. This allows the reader to grasp at a glance how
each idea relates to the others.
In the next chapter, we look at narrative pathway.
99
Notes
1. Garner B, Securities Disclosure in Plain English (Chicago: CCH Incorporated, 1999), 64.
2. Strunk, The Elements of Style, 15–17 (see chap. 4, n. 2).
3. Garner, Securities Disclosure in Plain, 61–73.
4. Caraco, “Genes and the Response,” 2868 (see chap. 2, n. 43).
5. Ramagnoli S, “Circulatory Failure: Exploring Macro-and Micro- Circulation,” Trends in
Anaesthesia and Critical Care 3 (2013), under “The Pulmonary Artery Catheter [PAC],” http://
www.trendsanaesthesiacriticalcare.com/article/S2210-8440(13)00020-8/fulltext.
6. Rayala, “Common Questions about Vasectomy,” 760 (see chap. 2, n. 7).
7. Iverson C, et al. ed. AMA Manual of Style, 10th ed. Oxford: Oxford University Press, 2007.
8. Booth W, Colomb G, Williams J, The Craft of Research, 3rd ed. (Chicago: University of Chicago
Press, 2008); see chap. 15 on “Communicating Evidence Visually.”
9. Slomski A, “Clinical Trials Update: Depression Remains Stable in Smokers Taking Varenicline
to Quit,” JAMA 310, no. 16 (2013): 1165.
10. “Scotland: Towards a Healthy and Interdependent Future,” Lancet 384, no. 9947 (2014),
http://www.thelancet.com/journals/;ancet/article/PIIS0140-6736(14)61614-7/fulltext.
100
101
CHAPTER 9
Choose a clear narrative pathway
If he would inform, he must advance regularly from Things known to
things unknown, distinctly without Confusion, and the lower he begins
the better. It is a common Fault in Writers, to allow their Reader too
much knowledge: they begin with that which should be in the Middle,
and skipping backwards and forwards, ‘tis impossible for anyone but he
who is perfect in the Subject before, to understand their Work, and such
an one has no Occasion to read it.—Benjamin Franklin1
Why does an expert in their field talk over everybody’s head? Why can’t they talk
about their subject in a simple way? Is it just because their ideas are too advanced
or complex? Perhaps. But sometimes when an author understands their subject
well, they become blind to what other people do or don’t understand. They over-
estimate the reader’s knowledge. They misjudge the widest reasonable audience,
defining it too narrowly, composed of people just like themselves. As a result, they
may do several things that confuse the reader.
101
102
talking about prior research and current knowledge and practice. For example,
the article, “Effect of Aspirin and Antiplatelet Drugs on the Outcome of the Fecal
Immunochemical Test,” begins by reviewing facts about colorectal cancer and
common screening techniques.
Originali Revised
DESCRIPTION DESCRIPTION OF THE ISSUE
Update of the US Preventive Services Should a doctor screen the vision of a
Task Force (USPSTF) recommendation patient over 65? (wseg = 11/11.0/72.6/5.8)
on screening for impaired visual acuity in
older adults. (wseg = 19/19.0/36.1/12.9)
RECOMMENDATION RECOMMENDATION (UPDATE)
The USPSTF concludes that the current None. We don’t think there is enough
evidence is insufficient to assess the evidence to decide this. (wseg = 11/5.5/
balance of benefits and harms of 85.8/2.6)
screening for impaired visual acuity in
older adults. (wseg = 26/26.0/37.2/14.5)
US Preventive Services Task Force, “Screening for Impaired Visual Acuity in Older Adults: US Preventive
i
We can make this statement clearer by revising it to state the issue in shorter,
real-world terms. Getting the first step right, by stating the issue in real-world
terms, helps make everything that follows easier to understand.
Though it’s usually best to start a narrative by talking about the real world,
there are a few exceptions. An article could start with a well-known abstract
concept and move to more concrete discussion. For example, an article could
start by taking about the theory of evolution, a well-known abstract concept,
and go on to talk about how it applies in immune therapy used to treat a patient
with HIV.
Does the narrative start by talking about the real world? Or does it seem
abstract?
reader, and stick to it. It often helps a reader follow a narrative if they know where
it is going. Table 9-2 shows examples of common narrative pathways.
How can you make your narrative pathway clear to the reader? It often takes
only a simple cue. Consider the paragraph heading: “How Common is Barrett
Esophagus and What Are the Risk Factors?”3 From this statement, the reader
learns what types of information will follow. But a reader might be confused if
this same paragraph went on to talk about the history of Barrett’s Esophagus or
ways to treat it.
separate sections for study design, patients or study subjects, study outcomes, and
interpretation of results. Within each section, it is often clear from context whether
the discussion deals with the real world or abstract ideas. Keeping real-world
and abstract ideas in separate sentences or paragraphs can also help make the
transition clear.
1. What real-world things and actions does the narrative mention? What abstract
ideas does it mention? Does the narrative make a smooth transition between
concrete and abstract?
2. In what way are a rat kidney and a human kidney enough alike that it makes
sense to compare them? Does the narrative explain this?
3. In what way are the kidneys of a rat, human, elephant and whale enough alike
we can generalize about an abstract “mammal kidney?” Does the narrative
explain this?
4. Write a paragraph that tells why we can generalize about a “mammal kidney.”
Compare the number of nephrons in a rat kidney and a human kidney.
Conclusion
In this chapter, we talked about choosing a clear narrative pathway. This usually
involves starting with things known and anchoring the discussion in the real
world. From there, choose a good narrative pathway, make it clear to the reader,
and stick to it. Make a smooth transition between concrete and abstract.
We develop these ideas further in the next chapter, where we give tips on forg-
ing a strong chain of logical reasoning.
Notes
1. Quoted in Williams, Style: Lessons in Clarity, 74 (see Preface, n. 8).
2. Bujamda, “Effect of Aspirin and,” 683 (see chap. 1, n. 27).
3. Zimmerman, “Common Questions about Barrett’s,” 92 (see chap. 2, n. 16).
106
╇ 107
CHAPTER 10
Forge a strong chain of logical reasoning
Looking back, I think it was more difficult to see what the problems were
than to solve them.—╉Charles Darwin1
Why doesn’t an expert write clearly? In this chapter, we give the second part of our
answer. An expert, someone “too close” to their subject, doesn’t always present a
strong chain of logical reasoning. Sometimes they leave out a step of reasoning
that seems obvious to them or present ideas in a confusing order (Table 10-╉1).
Many subjects would interest a wider audience, if only the author would explain
the subject step by step in the right order. In this chapter, we give tips on forging a
strong chain of logical reasoning.
107
108
The narrative starts by talking about the major parts of a kidney, the cortex
and medulla. It goes on to talk about a nephron and its parts.
W H AT ?
What is the problem? What is the question the research project is designed to
address? Most medical research sets out to answer a question (e.g., Does treat-
ment “A” work better than treatment “B?” Does the new medicine work better than the
old one?)
W HY ?
Why is the question or problem important? For a doctor or other medical scien-
tist, a problem becomes important when it affects the health or treatment of a
patient.
HO W ?
How did the research seek to answer the question or solve the problem? How was
the experiment or trial conducted?
It can be harder to follow a narrative that mixes up these three questions or
leaves one of them out. Some authors try to tell what they did and how they did it in
the same sentence. Some explain how they solved a problem without first stating
109
clearly, what the problem is, or why it is important. Some authors explain how
they carried out their research without first clearly stating what the question was
their research was designed to answer. A good statement of the research question
or problem goes a long way toward helping a reader understand the solution.
A medical research article starts with an abstract that addresses the what,
why and how questions. But an abstract doesn’t always frame these questions as
clearly as it might. For example, consider the statements of “Importance” and
“Objective” from a research article abstract in Table 10-2.
Do these statements answer the what and why questions as clearly as possible?
We can understand these statements with a little study. But we can revise to state
the research question, and tell why it is important, more simply and clearly. No
doubt the authors had to write their abstract to fit the journal’s format, which
required them to state their research “objective” rather than to frame the research
question.
The main text also addresses the research question, but clouds it with details of
research method and analysis.
Originali Revised
Importance Importance
Achieving glycemic control remains For a patient with type 2 diabetes, it can be
a challenge for patients with type 2 hard to keep blood sugar under control, even
diabetes even with insulin therapy. if they take insulin. (wseg = 22/22.0/65.2/9.6)
(wseg = 16/16.0/26.6/13.5)
Objective Objective—Issue addressed
To assess whether a fixed ratio of A patient has type 2 diabetes. Their doctor
insulin degludec/liraglutide was treats them using insulin glargine and
noninferior to continued titration metformin.
of insulin glargine in patients with If this doesn’t work to control their blood
uncontrolled type 2 diabetes treated sugar, the common next step is to raise their
with insulin glargine and metformin. insulin dose, as needed. This study poses the
(wseg = 31/31.0/3.4/20.4) question: Would it work just as well to use
a fixed ratio of insulin degludec/liraglutide?
(wseg = 56/14.0/67.2/7.3)
Lingvay I, et al. “Effect of Insulin Glargine Up-titration vs Insulin Degludec/Liraglutide on Glycated
i
Hemoglobin Levels in Patients With Uncontrolled Type 2 Diabetes—The DUAL V Randomized Clinical
Trial,” JAMA 315, no. 9 (2016): 898.
110
Part of what makes this excerpt hard to understand is that it covers both what
(the research question) and how (research method and analysis). The research
question is: Does degludec/liraglutide work as well as insulin glargine with metformin
to help control a patient’s diabetes?
The research method involves how the researchers designed the clinical trial,
and analyzed the data, in order to answer the research question. This involves talk
of: up-titration, baseline HbA1c level, statistically superior, body weight, and rate of
confirmed hypoglycemia. The statement of the research question would be clearer if
the discussion of what and how were kept separate.
The HMS Titanic sailed on its maiden voyage in April 1912 with 2223
passengers and crew aboard. The ship had 20 lifeboats, and each life-
boat could carry 60 people. If the ship were to strike an iceberg and
sink quickly, far from any source of help, how many people would have
to be left behind?
111
M AK I NG A S MO O T H T R A N SI T ION
safe, the TTC needs to be lower than, or equal to, the maximum safe
concentration (MSC). We can express this idea in the equation: MEC ≤
TTC ≤ MSC.3 (wseg = 88/14.6/65.0/7.8)
Conclusion
In this chapter, we gave tips to help you forge a strong chain of logical reason-
ing: Explain each step of reasoning. State the problem before you solve it. Say it in
words before you say it in symbols.
113
Notes
1. Quoted in Williams, Style: Lessons in Clarity, 185 (see Preface, n. 8).
2. Lingvay I, et al. “Effect of Insulin Glargine Up-titration vs Insulin Degludec/Liraglutide on
Glycated Hemoglobin Levels in Patients With Uncontrolled Type 2 Diabetes—The DUAL V
Randomized Clinical Trial,” JAMA 315, no. 9 (2016): 898–899.
3. Linares O, et al. “Personalized Oxycodone Dosing: Using Pharmacogenetic Testing and Clinical
Pharmacokinetics to Reduce Toxicity Risk and Increase Effectiveness,” Pain Med 15, no. 5 (2014).
114
115
In the 21st century, English is the global language of medical science. Reaching
the widest reasonable audience requires writing in a way that is understandable.
In this book, we showed how medicus incomprehensibilis mostly stems from
needless grammatical complexity. We identified several over-used writing habits
that are symptoms of this complexity. We showed how you can use a small collec-
tion of plain English writing tips to treat medicus incomprehensibilis and improve
reading ease, vividness and logical flow.
Change comes when people share a vision. We hope this book has provided
a clear vision of what plain English is, why it’s important, and how to use it to
improve your medical writing.
115
116
117
Appendix 1
Introduction
Anybody who wants to write for the widest reasonable audience needs to consider
the world’s non-native English speakers. This appendix surveys English speak-
ers around the world, including both native and non-native speakers. We pres-
ent data on English speakers in general, since it gives a rough idea of where the
world’s English-speaking doctors live.
Non-native English speakers constitute the majority of the world’s total English
speakers. In predominantly English-speaking countries, they are an important
minority. Since doctors are among the best-educated people in any country, it
stands to reason they are over-represented among each country’s English speak-
ers. For example, if 10% of India’s citizens speak English, it stands to reason far
more than 10% of Indian doctors speak English. (We think the number is closer to
100%.) The same probably holds true in any other country where learning English
is considered part of a “good education.”
117
118
Table A1-1. World’s top 25 English-speaking countries (population in millions)i
U S A , UK , C A NA DA A N D AU S T R A L I A
These large countries, where most people speak English, are home to 407 million
people or 5.5% of the world’s population.1 Within these countries, about 14% of
the total population, or 59 million people, are non-native English speakers. Taken
together, these countries have 247 medical schools (Table A1-2).
How many doctors in these countries are non- native English speakers?
Fourteen percent? More, or fewer? We don’t know, but we think the number is
fairly high.
These six large countries with old colonial ties to the UK or USA are home
to 1,900 million people, more than 25% of the world’s population. In these
E UROPE
Europe is home to about 740 million people, or about 11% of the world’s popula-
tion.2 European researchers publish widely in English-language journals. About
13% of Europeans speak English as a native language, mostly in the UK and
Ireland. Another 38%, about 191 million, speak English as an additional language.3
English is the most widely spoken foreign language in the European Union.
Overall, about 51% of the people of the European Union speak English.4 This rep-
resents a total population four times the size of the UK. In some countries, a large
majority of the people speak English, including the Netherlands (90%), Malta
(89%), Sweden (86%), Cyprus (73%), Austria (73%) and Finland (70%).5
Based on this information, we know many European doctors read English-
language medical journals. It also seems likely their level of medical science
knowledge exceeds their level of English-language skills.
CHINA
In China, 10 million people speak English, less than 1% of the population. This
number seems poised to grow in coming years, since China has another 300 mil-
lion “learners.”6 If just 5% of these “learners” successfully learn English, it would
add a new population of English speakers the size of the UK. (A 5% rate would be
about the same as in Brazil, where relatively few people speak English.)
We assume English-language medical literacy in China is low. As more Chinese
people learn English, we expect the number of Chinese doctors who read English
language medical journals to grow even faster. Plain English medical writing
would help to speed up this process.
Conclusion
Given the large and ever-increasing number of English speakers around the world,
it makes sense to write about medical science in plain English. Many research-
ers write their articles in English, though English is not their native language.
The data suggest non-native speakers now comprise a large part of the audience
123
for English-language medical journals. More doctors would read English language
journals, and understand them better, if they were written in plain English.
Notes
1. World population estimate of 7,429 million based on World Population Clock, (accessed June 13,
2016) http://www.worldometers.info/world-population/.
2. Wikipedia s.v. “Demographics of Europe,” (accessed June 12, 2016) https://en.wikipedia.org/
wiki/Demographics _of_Europe.
3. TNS Opinion & Social, Europeans and Their Languages, Special Eurobarometer 386
(Brussels: European Commission, June 2012), 5–6, 23.
4. Wikipedia, s.v. “Language in Europe,” (accessed June 21, 2016) https://en.wikipedia.org/wiki/
English_language_in _Europe.
5. TNS Opinion & Social, Europeans and Their Languages, Special Eurobarometer 386
(Brussels: European Commission, June 2012), 23.
6. Yang J, “Learners and users of English in China,” English Today 22, no. 2 (April 2006): 3–10.
124
125
Appendix 2
∂ S
(QCk ) = − J k
∂x L
∂ S
(Q ) = − J v
∂x L
125
126
∂
∂x
( )
QC pr = 0
where Q denotes plasma flow rate, S and L denote the surface area and length
of the capillary, Jv and Jk denote the fluid and solute fluxes, Ck denotes the total
plasma concentration (free and bound states) of solute k, the subscript pr denotes
protein, and x denotes the position along the capillary. Boundary conditions are
given for Q, Ck, and Cpr at the afferent end of the capillary. Volume flux is assumed
to be driven by hydrostatic and oncotic pressure differences, and fluxes for small
solutes (smaller than proteins) are assumed to be both advective and diffusive,
through the fenestrated capillary walls. (wseg = 320/20.0/41.4/11.8)
Notes
1. Layton A, “Mathematical Modeling of Kidney Transport,” NIH Public Access¸ under “2.
Glomerular Filtration,” http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3745785/; originally
published in Wiley Interdiscip Rev Syst Bio Med 5, no. 5 (September 2013).
2. Huynh T, et al. “The Frequency and Cost of Treatment Perceived to be Futile in Critical Care,”
JAMA Intern Med 173, no. 20 (2013): 1888.
3. Hanyaloglu A, von Zastrow M, “Regulation of GPCRs by Endocytic Membrane Trafficking and Its
Potential Implications,” Annual Review of Pharmacology and Toxicology 48 (2008): 538.
128
╇ 129
Appendix 3
EXERCISE KEY
Introduction
Exercise A.╇ Widest reasonable audience
1. We think the widest reasonable audience includes doctors, physiologists, phar-
macists, mathematicians, biologists, bio-╉medical engineers, bio-╉chemists, and
others. It might include researchers and advanced students in these fields.
2. A mathematician or engineer probably needs more information on kidney
anatomy. A doctor probably needs a better explanation of math and fluid
dynamics. For example, what is an ODE? (An ordinary differential equa-
tion?) Why explain kidney anatomy any doctor knows, but not technical jar-
gon related to fluid dynamics some doctors don’t know (e.g., afferent, efferent,
advective, oncotic, diffusive, fenestrated)?
2. We find the first excerpt, “Mathematical Modeling for Kidney Transport,” easi-
est to read. It uses shorter sentences and seems more concise than the others.
The second excerpt, on “Futile Critical Care,” uses longer sentences and seems
“wordy.” The third excerpt, “Regulation of GPCR’s by Endocytic Membrane
129
130
Trafficking,” is the hardest to read. The subject is very technical. It uses long
sentences and many long words.
3. Yes, there seems to be a general correlation between reading ease and sentence
length.
3. Given how complex the immune system is, any one B cell is
not likely to develop in a way that is easy to predict. For example,
a cell that reacts to an antigen expands to a clone that produces a
single antibody. More likely, the cell develops through a series of
error-prone, random events, in which it mutates and becomes re-
arranged. In the process, selective pressures determine whether the
cell still specifies the original antigen. (wseg = 75/18.7/52.4/10.6)
S UM M A RY
Comparing before and after wseg scores, we see reading ease improves by 49.0
points and the grade level drops by 11.1 grades (Table A3-3).
Original Revision
w s e g w s e g
Scores 72 36.0 3.4 21.7 75 18.7 52.4 10.6
Change 3 −17.3 49.0 −11.1
The revision has three sentences with 27, 1, and 22 words. The sentences have a
total of 50 words and an average sentence length of 16.6.
Revision:
Our Review focuses on new knowledge of COPD and its risk fac-
tors, prevalence, and history since these Reviews were published.
It addresses some questions that still persist. It also raises some
issues health-care planners must think about as the burden of
COPD increases as the world’s people age. (wseg = 48/16.0/61.9/8.5)
3. Since CYP2C9 explained 6 to 10 percent of the variability in
these two patient samples, the VKORC1 genotype appears to be the
most important genetic factor determining variability in warfa-
rin dose: in both clinical populations its effect was approximately
three times that of the CYP2C9 genotype. (wseg = 46/46.0/1.9/24.4)
Revision:
Revision:
Revision:
Over the past 100 years, exercise science has grown. It began with
work showing how exercise affects vascular control, heat produc-
tion, oxygen need, and lactic acid dynamics. This led to a Nobel Prize
in physiology or medicine for August Krogh of Denmark in 1920.
It led to another one for A.V. Hill of the United Kingdom and Otto
Meyerhof of Germany in 1922. Since then other studies have led to
our modern knowledge of fitness. We now know fitness, measured
by a person’s peak rate of oxygen use, strongly predicts their chance
of illness or death. (wseg = 96/16.0/53.1/9.8)
6. In a prospective study in the Netherlands that followed more
than 30,000 students 10 to 14 years of age for up to three years,
annual scoliosis screening in addition to the usual biennial health
checkup detected no cases of idiopathic scoliosis requiring surgery,
and the authors concluded that additional annual scoliosis screen-
ing was not needed. (wseg = 55/55.0/0.0/ 28.6)
Revision:
S U MM A RY
This exercise asked you to practice using normal sentence length and making
other changes to improve reading ease. Table A3-4 summarizes the wseg scores
for the original sentences and our revisions. On average, reading ease improved
by 50.8 points and grade level dropped 19.3.
Original Revision
w s e g w s e g
Exercise 1.B. Keep the subject and verb close together in the first seven or
eight words
1. Although SDM is well accepted in overtly value-laden clinical
decisions such as prostate-specific antigen testing and mammog-
raphy screening, the principles of SDM apply to a broad range of
health care decisions, discharges against medical advice included.
(wseg = 36/36.0/0.0/22.3)
Only two words separate the subject and the verb, but they do not come within
the first eight words.
135
Revision:
Fourteen words separate the subject, effort, and the verb, is. The verb does not
come in the first eight words.
Revision:
In this revision, no words separate the subjects and verbs. The long name, “Global
Burden of Diseases, Injuries, and Risk Factors (GBD),” prevents us from putting the
subject, enterprise, within the first eight words.
Only four words separate the subject and the verb, but they do not appear within
the first eight words.
Revision:
the reliability of both the review and the individual studies and to
assess their relevance to clinical practice and the meaning of the
results reported in the review. (wseg = 57/57.0/0.0/28.7)
Nineteen words separate the first subject, reporting, and the verb, are. Thirteen
words separate the second subject, reporting, and the verb, are. The verb does not
come in the first eight words.
Revision:
Twenty-six words separate the first subject, characteristics, and the verb, are
presented. Twenty-three words separate the second subject, distribution, and the
verb, are presented. The verb does not come in the first eight words.
Revision:
In this example, the subject and verb are together, but they don’t come within the
first eight words of the sentence.
137
Revision:
Some patients don’t want to talk about a traumatic event, and may
avoid treatment as a result. That’s why it’s important to talk with
them about their preferences for treatment. (wseg= 30/15.0/67.5/7.5)
S UM M A RY
This exercise asked you to practice keeping the subject and the verb close
together in the first seven or eight words. Table A3-5 shows the wseg scores for
our revisions.
Original Revision
w s e g w s e g
1 36 36.0 0.0 22.3 36 12.0 48.9 9.4
2 31 31.0 28.0 17.0 35 17.5 48.8 10.7
3 39 39.0 2.3 22.6 34 11.3 58.4 7.9
4 57 57.0 0.0 28.7 58 14.5 55.0 9.1
5 34 34.0 0.0 23.6 33 11.0 57.2 8.0
6 26 26.0 21.0 16.7 30 15.0 67.5 7.5
Average 37.2 37.2 8.6 21.8 37.7 13.6 56.0 8.8
Change 0.5 −23.6 47.4 −13.1
Our first task was to estimate mortality for certain age ranges
for the year 2000. We started by reviewing data we had on births,
deaths and census. Then we corrected for under-reported deaths
using the synthetic extinct generation and growth balance methods.
We assessed under-5 mortality or 5q0, and age 15-to-60 mortality
or 45q15. We defined “under-5 mortality” as “the chance of death
between 0 and 5 years of age.” We defined “age 15 to 60 mortality”
as “the chance of death between 15 and 60 years of age.” (wseg = 89/
14.8/52.0/9.6)
3. The trial showed that pharmacogenetic- guided initiation of
warfarin therapy resulted in a greater percentage of time in the
therapeutic range, fewer excessive INRs, a shorter median time to
therapeutic INR, and fewer dose adjustments. (wseg = 34/34.0/0.6/
21.6)
This sentence already puts the main point first. But we can make it more concise,
and move the supporting details to a second sentence.
Revision:
We think there are two key points. One comes at the start, the other, 47 words
into the sentence. We revised to put each key point at the start of its own
sentence.
Revision:
The participation rate was lower than we planned when we did the
power calculations. But we doubt that a participation rate of 70%
would have made any difference. Though more people than expected
had increased risk and received counselling, we saw no trend towards
a decrease in ischaemic heart disease. (wseg = 50/16.6/56.2/9.5)
5. Recommended CRC screening strategies fall in 2 broad catego-
ries: stool tests that primarily detect cancer, which include detec-
tion of occult blood or exfoliated DNA, and structural tests, such
as flexible sigmoidoscopy, colonoscopy, and computed tomographic
colonography, which are effective in detecting both cancer and pre-
malignant lesions. (wseg = 46/46.0/0.0/27.4)
This sentence spreads out the main point. You need to read 27 words to get the
whole thing.
Revision:
We think the main point is broken up throughout the sentence. You don’t get the
whole main point until the end. In the revision, we broke up this long sentence to
make a short paragraph.
Revision:
S U MM A RY
This exercise asked you to practice putting the main point first before giv-
ing commentary, detail or support. Table A3-6 shows the wseg scores for our
revisions.
Original Revision
w s e g w s e g
1 44 44 16 21.9 51 12.7 62.8 7.6
2 77 77 0.0 36.6 89 14.8 52.0 9.6
3 34 34 0.6 21.6 35 17.5 41.6 11.8
4 60 60 38.2 15.3 50 16.6 56.2 9.5
5 46 46 0.0 27.4 62 15.5 58.7 8.9
6 30 30 0.0 22.8 40 13.3 49.4 9.6
Average 48.5 48.5 9.1 24.3 54.5 15.1 53.5 9.5
Change 6.0 −33.4 44.3 −14.8
Revision:
Revision:
Any Global Burden of Disease Study must assess data with care. In
this assessment, the sum of deaths by specific cause, age and sex
must equal the sum of deaths from all causes assessed separately.
(wseg = 35/17.5/63.3/8.7)
Revision:
Revision:
For a patient to get good treatment, the doctor needs to make the
right diagnosis. (wseg = 15/15.0/73.1/6.7)
5. The 2 objectives of the study were (1) to examine the association
between physical activity and dietary behavior and (2) to exam-
ine the potential combined effect of physical activity and dietary
behavior on biological (eg, total cholesterol) and health (eg, waist
circumference) markers. (wseg = 43/43.0/0.0/25.8)
142
Revision:
Our study checked the link between diet and exercise, and their
combined effect on bio-and health-markers (eg, total cholesterol
and waistline.) (wseg = 23/23.0/51.0/11.8)
6. Assuming vasal disruption and occlusion have been adequately
achieved during surgery, and assuming the patient adheres to using
another contraceptive method while awaiting confirmation of ste-
rility, true causes of vasectomy failure include recanalization (early
and late) and, more rarely, aberrant anatomy (e.g., the presence of a
third vas). (wseg = 48/48.0/0.0/27.7)
Revision:
Why does a vasectomy fail? The surgery may fail to disrupt and
occlude the vas. A patient may fail to use other birth control while
he waits for test results to prove he is sterile. The vas may re-cana-
lize sooner or later. Or a patient may have odd anatomy, such as a
third vas. (wseg = 54/10.8/78.3/5.0)
S U MM A RY
This exercise asked you to keep essential scientific terms but minimize other long
words. Table A3-7 shows the wseg scores for our revisions.
Original Revision
W S E G W S E G
1 26 26.0 11.2 18.1 31 15.5 62.8 8.3
2 51 51.0 0.0 26.7 35 17.5 63.3 8.7
3 40 40.0 34.2 13.4 32 10.6 50.6 8.8
4 8 8.0 0.0 17.0 15 15.0 73.1 6.7
5 43 43.0 0.0 25.8 23 23.0 51.0 11.8
6 48 48.0 0.0 27.7 54 10.8 78.3 5.0
Average 36.0 36.0 7.6 21.5 31.7 15.4 63.2 8.2
Change −4.3 −20.6 55.6 −13.2
143
One new study shows dosing guided by genotype gives better con-
trol of anti-coagulation. But this finding was based on comparing
144
Bio-active, macro-nutrient.
Revision:
Multi-modal.
Revision:
S U MM A RY
This exercise asked you to practice writing a compound word to maximize reading
ease and show how you pronounce it. Table A3-8 shows the wseg scores for our
revisions.
145
Original Revision
w s e g w s e g
1 27 27.0 7.0 18.9 32 16.0 55.7 9.4
2 55 55.0 3.3 26.4 48 12.0 60.7 7.7
3 28 28.0 6.1 19.3 27 13.5 45.8 10.2
4 25 25.0 0.0 19.6 23 7.6 59.2 6.8
5 12 12.0 39.5 10.7 9 9.0 56.7 7.5
6 12 12.0 0.0 26.4 14 14.0 0.0 20.2
Average 26.5 26.5 9.3 20.2 25.5 12.0 46.4 10.3
Change −1.0 −14.5 37.0 −9.9
We think poor nutrition or poor nutrition status works without the –al ending. You
can also say polluted air rather than pollutants. We replaced respiratory with lung.
Revision:
But being poor is a proxy for factors that raise the risk for COPD.
Poor people tend to live in crowded quarters with poor nutrition
and poor access to health care. They breathe polluted air at work.
146
In low and middle income countries, many tend to smoke. All this
helps cause early lung infection. (wseg = 54/10.8/84.6/4.1)
3. The concept of pharmacogenetics originated from the clinical
observation that there were patients with very high or very low
plasma or urinary drug concentrations, followed by the realization
that the biochemical traits leading to this variation were inherited.
(wseg = 38/8.0/0.0/23.4)
We think urine drug concentration is good modern English. The words, clinical, and
chemical are used in common speech. But in the revision, we revised to avoid using
biochemical.
Revision:
We think exam and mammogram are good modern English. We would say,
age 50–64.
Revision:
The ending, familial is unnecessary, since you can just say, family history of. We
think gene testing or gene counselor would sound awkward. Referral is plain English,
but refer is shorter.
Revision:
S UM M A RY
This exercise asked you to practice dropping any unneeded word ending. Table A3-9
shows the wseg scores for our revisions.
Original Revision
w s e g w s e g
1 14 14.0 29.4 12.6 14 14.0 41.5 10.9
2 51 51.0 4.1 25.3 54 10.8 84.6 4.1
3 38 38.0 0.0 23.4 36 12.0 63.0 7.4
4 21 21.0 4.2 17.8 27 13.5 58.3 8.4
5 21 21.0 4.2 17.8 27 13.5 61.5 8.0
6 21 21.0 32.4 13.9 23 11.5 59.0 7.8
Average 23 23.2 13 18 30 12.6 61.3 7.8
Change 7.0 −10.7 48.0 −9.9
Revision:
Revision:
Revision:
Revision:
Revision:
Revision:
Tick paralysis results from the bite of a pregnant female tick. Tick
paralysis is a kind of ascending paralysis caused by a toxin. It gen-
erally resolves after the tick is removed. (wseg = 31/10.3/54.4/8.2)
S UM M A RY
This exercise asked you to practice identifying and eliminating noun strings. Table
A3-10 shows the wseg scores for our revisions.
150
Original Revision
w s e g w s e g
1 31 31.0 0.7 20.8 32 16.0 55.7 9.4
2 41 41.0 8.4 22.2 58 11.6 63.7 7.2
3 14 14.0 0.0 18.5 16 16.0 58.4 9.0
4 51 51.0 20.0 16.7 57 14.2 57.3 8.8
5 37 37.0 41.7 12.0 36 18.0 56.0 8.4
6 19 19.0 4.9 17.2 31 10.3 54.4 8.2
Average 32.2 32.2 12.6 17.9 38.3 14.4 57.6 8.5
Change 6.2 −17.8 45.0 −9.4
We think “and” sounds okay. In our revision, we didn’t use any conjunction to
replace furthermore. We did use the conjunction therefore as a way to break up the
long sentences.
Revision:
However, the gain in the past 40 years was only 11 years, compared
with total improvements of two to three times more in other parts
of Asia (eg, the Maldives), the Middle East (especially Oman), and
Latin American (eg, Bolivia, Peru, and Guatemala). (wseg = 72/24.0/
17.9/16.7)
Since 1970, the mean life span has changed unevenly. In 2010, a
woman in Japan could expect to live 85.9 years and likely still lon-
ger in 2012. But the gain in the past 40 years was only 11 years.
The gain was two to three times more in other parts of Asia (eg,
the Maldives), the Middle East (eg, Oman), and Latin America (eg,
Bolivia, Peru, and Guatemala). (wseg = 68/17.0/55.2/9.7)
3. Our analysis does not address the issue of whether a precise ini-
tial dose of warfarin translates into improved clinical end points,
such as a reduction in the time needed to achieve a stable thera-
peutic INR, fewer INRs that are out of range, and a reduced inci-
dence of bleeding or thromboembolic events. However, our study
lays important groundwork for a prospective trial and suggests
that such a trial should be powered to detect the benefits of incor-
porating pharmacogenetic information into the dose algorithm for
patients who require high or low doses—the subgroups in our study
for whom dose estimates based on the pharmacogenetic algorithm
differed significantly from those based on the clinical algorithm.
(wseg = 113/56.5/0.5/27.2)
These two long sentences draw a contrast between what the analysis does not
address and what it lays the groundwork for. We think just inserting the word but
would sound awkward since the contrasting ideas are so remote from each other.
We re-phrased to put these ideas closer together, and used but in the middle of
the sentence.
Revision:
This deals with several issues: Who should field each call? Who can
best deal with each problem? And who can best recognize each prob-
lem? But participants noted several drawbacks to this approach.
They included lack of subspecialty specificity, more effort without
more pay or overhead expense, and possible overuse of the system.
(wseg = 52/10.4/56.3/7.9)
6. This study found that older patients who had severe neurocar-
diogenic syncope (average of seven syncopal episodes in the pre-
vious two years and asystolic pauses averaging 11 seconds) had
a decreased time to first syncopal event after a pacemaker was
153
We think but would sound okay. In our revisions, we broke up the long sentence
into shorter ones and decided not to use any conjunctions.
Revision:
This study looked at older patients who had severe fainting caused
by brief heart pauses. Study subjects had an average of seven faint-
ing spells in the prior two years. During these spells, their hearts
paused for an average of 11 seconds. After they had a pacemaker
implanted, they had less time to a first fainting spell. This study was
a manufacturer-funded trial. It did not report the total burden of
fainting or number of falls. (wseg = 75/12.5/66.6/7.0)
S UM M A RY
This exercise asked you to practice replacing a long conjunction at the beginning
of a sentence. Table A3-11 shows the wseg scores for our revisions.
Original Revision
w s e g w s e g
1 58 29.0 27.1 16.6 59 14.7 61.3 8.3
2 72 24.0 17.9 16.7 68 17.0 55.2 9.7
3 113 56.5 0.5 27.2 105 15.0 55.4 9.2
4 59 29.5 0.0 20.9 79 13.1 51.0 9.4
5 43 21.5 0.0 21.6 52 10.4 56.3 7.9
6 59 29.5 13.4 18.7 75 12.5 66.6 7.0
Average 67.3 31.7 9.8 20.3 73.0 13.8 57.6 8.6
Change 5.7 −17.9 47.8 −11.7
S UM M A RY
This exercise asked you to practice identifying long words and replacing or elimi-
nating them. Table A3-12 summarizes the changes.
Our revision uses 29 long words and 262 total words (29 long words/262
total words = 11.1%). Overall, we reduced the number of long words by 55% and
improved the reading ease score by 30.0 points.
156
HO W DI D W E D O I T ?
We kept the 13 words we identified as essential scientific terms and used 16 other
long words. We replaced some long words with their plain-English equivalents.
Thus, analgesic becomes pain relief. Respiratory becomes breathing. Miotic becomes
pupil function.
We replaced other long words with shorter words: majority, nevertheless, affin-
ity, experience, exaggerated, similar, albeit, dramatic, illustrate, relevant, mediated,
virtually, observations, perspective, relevant, associated, markedly, and attenuated.
We also changed some nominalizations into verbs in root form. Thus, con-
version becomes converts. Metabolism becomes metabolizes. (We talk more about
nominalization in Chapter 4.)
We also broke the one paragraph into five paragraphs to reflect what we
thought was the natural progression of ideas.
We struck nine words and added one for a net of eight words (8/42 = 9.0%).
Revision:
We struck nine words and added two, for a net of seven (7/32 = 22.9%).
Revision:
New direct acting antiviral drugs are causing a big change in the
treatment for hepatitis C virus (HCV). (wseg = 18/18.0/56.9/9.7)
5. Both dietary and exercise physical activity behavior are inde-
pendent predictors of numerous health outcomes among adults.
(wseg = 15/15.0/0.0/17.7)
We struck six words and added one, for a net of five (5/15 = 33.3%).
Revision:
S U MM A RY
This exercise asked you to practice deleting needless words. Table A3-13 shows the
wseg scores for our revisions.
Original Revision
w s e g w s e g
1 28 28.0 15.2 18.0 30 15.0 73.1 6.7
2 42 42.0 29.2 19.6 35 17.5 56.1 9.7
3 31 31.0 0.7 20.8 42 10.5 49.1 9.0
4 32 32.0 21.0 18.2 18 18.0 56.9 9.7
5 15 15.0 0.0 17.7 10 10.0 61.3 7.1
6 168 21.0 22.3 15.3 81 13.5 63.6 7.7
Average 52.7 28.2 14.7 18.3 36.0 14.1 60.0 8.3
Change −16.7 −14.1 45.3 −10.0
159
Revision:
Revision:
The UNPD’s estimates of deaths for children under age 5 are much
higher than ours. For the years 2005 through 2010, their estimates
are 8 million higher. (This is an extra 1 to 6 million deaths per year.)
(wseg = 38/12.6/67.0/7.0)
3. Recurring themes in pharmacogenetics include the presence of a few
relatively common variant alleles of genes encoding proteins important
in drug response, a larger number of much less frequent variant alleles,
and striking differences in the types and frequencies of alleles among
different populations and ethnic groups. (wseg = 47/47.0/0.7/24.8)
Revision:
Revision:
The quality of the studies varied a lot. Many were old. Only a few
gave enough detail so we could copy the intervention. (wseg = 23/
7.6/70.3/5.3)
160
5. At the time that the 2004 algorithm was published, there were 2
available rigorous evidence-base reviews of the treatment of RLS/
WED prepared under the auspices of the Standards of Practice
Committee of the American Academy of Sleep Medicine. (wseg = 39/
39.0/13.2/21.1)
Revision:
Revision:
S U MM A RY
This exercise asked you practice omitting any needless of. Table A3-14 shows the
wseg scores for our revisions.
Original Revision
w s e g w s e g
1 36 36.0 19.8 19.4 31 15.5 49.1 10.2
2 35 35.0 34.3 12.8 38 12.6 67.0 7.0
3 47 47.0 0.7 24.8 44 11.0 49.5 9.0
4 25 25.0 35.0 11.4 23 7.6 70.3 5.3
5 39 39.0 13.2 21.1 36 18.0 23.7 13.7
6 16 16.0 47.5 10.0 17 17.0 55.2 9.7
Average 33.0 33.0 25.1 16.6 31.5 13.6 52.5 9.2
Change −1.5 −19.4 27.4 −7.4
161
Revision:
Revision:
Revision:
Revision:
More recent work by Goyal et al shows high blood sugar at or after
admission predicts 30-day death rate for patients with AMI. For this
purpose, high blood sugar was defined as blood glucose ≤ 3.8 mmo/
L. (wseg = 38/19.0/69.5/8.2)
6. Overt hyperthyroidism that is inadequately treated is associated
with an increased risk of adverse maternal and neonatal outcomes
(Table 4). (wseg = 20/20.0/8.8/16.9)
Revision:
S U MM A RY
This exercise had you practice omitting the needless that. Table A3-15 shows our
wseg scores.
Original Revision
w s e g w s e g
1 21 21.0 0.2 18.4 14 14.0 33.6 12.2
2 19 19.0 53.9 10.4 17 17.0 50.2 10.4
3 41 41.0 6.3 22.5 38 12.6 53.7 8.9
4 19 19.0 53.9 10.4 15 15.0 56.2 9.1
5 31 31.0 22.5 17.8 38 19.0 69.5 8.2
6 20 20.0 8.8 16.9 23 11.5 66.4 6.8
Average 25.2 25.2 24.3 16.1 24.2 14.9 54.9 9.3
Change −1.0 −10.3 30.7 −6.8
163
3. Warfarin binds to albumin, and only about 3% is free and phar-
macologically active.
The first clause is active; the second is neither active nor passive.
4. After entry, the 9.6 kb viral genome undergoes cytoplasmic trans-
lation into a single polypeptide, which is subsequently {cleaved}
into 10 viral proteins—three structural and seven non-structural.
The main clause’s grammatical form looks active; but it seems passive, since
the genome isn’t acting upon anything (or upon itself). The second clause is
passive.
We collected 24-hour urine samples each day for three days and
checked for volume and sodium. (wseg = 17/17.0/65.1/8.3)
2. COPD can be {classified} with respect to both phenotype and dis-
ease severity. (wseg = 12/12.0/46.6/9.7)
This sentence’s main clause is passive. Some of the other clauses are active.
Revision:
5. The incidence of major injury in each of the cohorts was {calcu-
lated} per 10,000 person-years. (wseg = 15/15.0/11.1/15.4)
Revision:
For each cohort, we computed the rate of major injuries. We gave this
rate in terms of per 10,000 person years. (wseg = 21/10.5/63.2/7.0)
6. The exact pathophysiologic mechanism for scoliosis is {unknown}.
(wseg = 8/8.0/0.0/17.0)
S UM M A RY
This exercise asked you to practice revising a passive sentence into active voice.
Table A3-16 shows the wseg scores for our revisions.
Original Revision
w s e g w s e g
1 16 16.0 26.6 13.5 17 17.0 65.1 8.3
2 12 12.0 46.6 9.7 11 11.0 72.6 5.8
3 31 31.0 30.7 16.6 33 16.5 61.9 8.7
4 Sentence Active
5 15 15.0 11.1 15.4 21 10.5 63.2 7.0
6 8 8.0 0.0 17.0 7 7.0 42.6 9.0
Average 16.4 16.4 23.0 14.4 17.8 12.4 61.1 7.8
Change 1.4 −4.0 38.1 −6.7
Revision:
Revision:
You can’t make this assessment just by adding reported causes. Why
is this? Since there are likely more data reported for the death rate
from all causes than for individual causes. You need to assess age-
specific death rate separately. This helps double check against the
less robust estimates of death rate from specific causes for a group
defined by age and sex. (wseg = 62/12.4/65.9/7.1)
3. The current hospitalist-ambulist division of general medical care
has made important contributions to patient care, but it leaves
much to be {desired}, especially with regard to personalization and
continuity of care. (wseg = 31/31.0/3.4/20.4).
Revision:
Revision:
Revision:
The guidelines for diagnosing and treating asthma first came out in
1991. Yet inhaled cortico steroids (ICSs) played a key role in treat-
ing asthma before then. (wseg = 26/13.0/50.4/9.4)
6. The cremasteric reflex, which is {elicited} by pinching the medial
thigh, causes elevation of the testicle. (wseg = 16/16.0/21.3/14.2)
Revision:
The cremasteric reflex lifts the testicle. A doctor can check this
reflex by pinching the inner thigh. (wseg = 17/8.5/63.8/6.4)
S UM M A RY
This exercise asked you to practice eliminating forms of to be and to have. Table
A3-17 shows the wseg scores for our revisions.
Original Revision
w s e g w s e g
1 11 11.0 72.6 5.8 10 10.0 69.7 6.0
2 58 29.0 14.0 18.5 62 12.4 65.9 7.1
3 31 31.0 3.4 20.4 40 13.3 60.0 8.1
4 26 26.0 30.7 15.4 23 23.0 69.4 9.2
5 29 29.0 11.1 18.9 26 13.0 50.4 9.4
6 16 16.0 21.3 14.2 17 8.5 63.8 6.4
Average 28.5 23.7 25.5 15.5 29.7 13.4 63.2 7.7
Change 1.2 −10.3 37.7 −7.8
168
1. To curb such empirical use, a report from the Infectious Diseases
Society of America (IDSA) is calling for steps to boost the devel-
opment of better diagnostic tests, to reduce regulatory hurdles
for new tests, and to improve clinical use of infectious disease
diagnostics.
2. In sub-Saharan Africa and southeast Asia, peer or community
counselling and condom distribution among female sex workers was
estimated to be cost effective, at US$86 per infection averted and
$5 per DALY averted (all costs from here expressed in 2012 US$),
and was more cost-effective than school-based education, voluntary
counselling and testing, prevention of mother-to-child transmis-
sions, and STI treatment.
3. We hypothesized that the administration of fixed-duration anti-
biotic therapy (4 days) after source control would lead to equivalent
outcomes and a shorter duration of therapy as compared with the
traditional strategy of administration of antibiotics until 2 days
after the resolution of the physiological abnormalities related
to SIRS.
4. Over 93% of participants in the control arm aged 40–49 returned
their annual questionnaire, whereas compliance with annual breast
examination screening for those in the control arm aged 50–59 var-
ied between 89% (for screen 2) and 85% (for screen 5); only ques-
tionnaires were obtained for 3% to 7% of the women.
5. For many of these reasons, evidence-based reviews generally
make authoritative statements on the degree of evidence in sup-
port of the use of each medication for a defined disorder, but they
are not always conducive to the development of practical algorithms
for the management of disorders of varying severity and a lengthy
natural history.
6. If history or examination findings raise concern for intracranial
lesions, magnetic resonance imaging of the brain can be useful for
further evaluation, with particular scrutiny of the skull base.
S U MM A RY
Revision:
Revision:
Revision:
Revision:
Revision:
Revision:
In 2011, the U.S. Preventative Services Task Force found too little
evidence to recommend for or against bladder cancer screening for
an adult with no symptoms. (wseg = 26/26.0/37.2/14.5)
S U MM A RY
This exercise asked you to practice replacing nominalization with a verb in active
voice. Table A3-18 shows the wseg scores for our revisions.
Original Revision
w s e g w s e g
1 30 30.0 0.0 21.6 29 14.5 25.8 13.2
2 32 32.0 47.4 14.5 41 13.6 67.0 7.2
3 15 15.0 44.9 10.7 13 13.0 43.9 10.3
4 35 35.0 16.6 19.6 24 12.0 50.1 9.2
5 12 12.0 67.7 6.7 7 7.0 66.7 5.6
6 26 26.0 11.2 18.1 26 26.0 37.2 14.5
Average 25.0 25.0 31.3 15.2 23.3 14.4 48.5 10.0
Change −1.7 −10.7 17.2 −5.2
171
How old is a woman when she has her first baby? In 2009, the mean
age was 25.2 in the USA and 30 in Germany and Britain. The mean
age has increased steadily over the past several decades. (wseg = 38/
12.6/64.8/7.3)
2. Fifth, the airflow limitation or obstruction that happens in
COPD is caused by a mixture of small airway disease, parenchymal
destruction (emphysema), and, in many cases, increased airways
responsiveness (asthma). (wseg = 30/30.0/15.6/18.5)
The subject airflow limitation or obstruction is concrete. The word airflow sounds
real-world, but the long, Latin-origin words, limitation and obstruction sound
abstract. The words parenchymal destruction and responsiveness also sound
abstract.
Revision:
Fifth, the airflow blockage with COPD may have a few causes.
There may be a mixture of small airway disease, lung tissue damage
(emphysema), and, in many cases, other airway narrowing (asthma).
(wseg = 32/16.0/58.4/9.0)
3. The goal of therapy is to keep the INR in the therapeutic range,
since patients with an INR that is subtherapeutic are at increased risk
for thrombosis and patients with an INR that is supratherapeutic are
at increased risk for bleeding. (wseg = 41/41.0/37.2/18.2)
Ischaemic heart disease is a real-world problem. However, the words morbidity and
mortality sound abstract.
173
Revision:
Over the past 75 years, more and more US women have been receiv-
ing prenatal care. (wseg = 15/15.0/67.5/7.5)
S UM M A RY
This exercise asked you to tell whether a subject is abstract or concrete, and replace
an abstract subject with a concrete subject. Table A3-19 shows the wseg scores for
our revisions.
Original Revision
w s e g w s e g
1 33 33.0 22.0 18.3 38 12.6 64.8 7.5
2 30 30.0 15.6 18.5 32 16.0 58.4 9.0
3 41 41.0 37.2 18.2 43 14.3 68.3 7.2
4 12 12.0 25.4 12.6 12 12.0 60.7 7.7
5 10 10.0 0.0 20.1 16 16.0 63.6 8.3
6 16 16.0 47.8 10.5 15 15.0 67.5 7.5
Average 23.7 23.7 24.7 16.4 26.0 14.3 63.9 7.9
Change 2.3 −9.4 39.2 −8.5
174
We made a time series of the death rate for each year from 1970
to 2010. It covered each age group in 187 countries and included
uncertainty. We did this by using the advances we noted above,
and by extending the Brass relational model life tables. (wseg = 46/
15.3/55.1/9.3)
One early study showed how gene variation affects the way the body
metabolizes a drug. The study checked isoniazid in plasma for a sub-
ject with a fast or slow rate of acetylation. The results showed two
distinct modes (Fig. 2). (wseg = 40/13.3/57.9/8.4)
4. There was evidence of bias when primary studies did not pro-
vide an adequate description of either the diagnostic (index) test or
the patients, when different reference tests were used for positive
and negative index tests, or when a case-control design was used.
(wseg = 42/42.0/19.1/21.0)
We found some bias in the primary studies. Some did not describe
the diagnostic (index) test or the patients well enough. Sometimes,
a study used one reference test for a positive index test and another
for a negative one. Some studies used a case-control design. (wseg =
45/11.2/58.1/7.9)
Original Revised
Adjectives & Total words Percent Adjectives & Total words Percent
adverbs adverbs
1 11 56 20% 9 56 16%
2 9 34 26% 7 46 15%
3 11 39 28% 5 40 13%
4 10 42 24% 9 45 20%
5 9 29 31% 5 42 12%
6 3 8 38% 1 11 9%
Total 53 208 25% 36 240 15%
Original Revised
w s e g w s e g
1 56 56.0 0.4 27.1 56 11.2 65.5 6.8
2 34 34.0 3.1 21.2 46 15.3 55.1 9.3
3 39 39.0 0.0 24.1 40 13.3 57.9 8.4
4 42 42.0 19.1 21.0 45 11.2 58.1 7.9
5 29 29.0 0.0 20.5 42 14.0 53.6 9.2
6 8 8.0 71.8 5.2 11 11.0 95.6 2.6
Average 34.7 34.7 15.7 19.9 40.0 12.7 64.3 7.4
Change 5.3 −22.0 48.6 −12.5
177
S UM M A RY
This exercise asked you to practice using nouns and verbs to carry the weight of
meaning. Table A3-20 shows we reduced adjectives and adverbs to 15%.
Table A3-21 shows wseg scores for our revisions.
Revision:
Revision:
Revision:
The same dose of a drug may cause one patient to have an adverse
effect and another to have no therapeutic response. (wseg = 22/
22.0/65.2/9.6)
4. Participants were referred to their general practitioner for medi-
cal treatment, if relevant. (wseg = 12/12.0/11.3/14.6)
178
Revision:
We referred each test subject to his or her own doctor for medical
treatment, if needed. (wseg = 16/16.0/68.9/7.6)
5. All patients had angiographically defined CAD with at least 1
vessel that met the American College of Cardiology/American Heart
Association (AHA/ACC) class I or II indications for PCI, and only
those who received implants with drug-eluting stents were consid-
ered eligible for the study. (wseg = 45/45.0/3.2/23.9)
Revision:
Each patient had CAD, as defined by heart imaging. They also had
at least one vessel that met the American College of Cardiology/
American Heart Assoc (ACC/AHA) class I or II indications for PCI.
Only those who received an implant with a drug-eluting stent took
part in the study. (wseg = 50/16.6/54.5/9.7)
6. Tinnitus occurs in most persons with normal hearing who are
exposed to silence. (wseg = 13/13.0/56.9/8.5)
Revision:
Original Revision
w s e g w s e g
1 18 18.0 9.9 16.3 27 13.5 67.7 7.1
2 38 38.0 0.0 26.2 38 9.5 50.2 8.6
3 25 25.0 19.0 16.8 22 22.0 65.2 9.6
4 12 12.0 11.3 14.6 16 16.0 68.9 7.6
5 45 45.0 3.2 23.9 50 16.6 54.5 9.7
6 13 13.0 56.9 8.5 14 14.0 53.6 9.2
Average 25.2 25.2 16.7 17.7 27.8 15.3 60.0 8.6
Change 2.7 −9.9 43.3 −9.1
179
S UM M A RY
This exercise asked you to practice writing in the singular. Table A3-22 shows the
wseg scores for our revisions.
Revision:
Each study patient had septic shock and high risk of death. Generally,
once we stabilized a patient’s blood pressure, our open-label use of
esmolol kept their heart rate within the target range of 80 to 94/
min. (wseg = 37/18.5/60.0/9.4)
2. Use of lung function to characterize severity is, currently, the
best system available to clinicians, but it clearly falls well short of
being ideal. (wseg = 24/24.0/37.9/13.9)
Revision:
Though less than ideal, testing lung function is still the best way
for a doctor to judge how severe a patient’s COPD is. (wseg = 23/
23.0/76.8/8.2)
3. The response to many drugs in common use varies greatly among
patients. (wseg = 12/12.0/60.7/7.7)
Revision:
For many drugs in common use, the response varies greatly from
patient to patient. (wseg = 14/14.0/65.7/7.5)
4. These agents seem to facilitate the use of shortened courses of
combination interferon- free therapy, which are associated with
high (>95%) sustained response rates and relatively few toxicities.
(wseg = 27/27.0/10.2/18.5)
Revision:
5. Lactate levels have become a useful marker for tissue hypoperfu-
sion and may also serve as an end point for resuscitation in patients
with sepsis and septic shock. (wseg = 27/27.0/35.2/15.0)
Revision:
A patient’s lactate levels are a useful marker for lack of blood flow
to tissue. They may also serve as an end point for reviving a patient
with sepsis or septic shock. (wseg = 32/16.0/71.6/7.2)
6. Women with hypothyroidism should be counseled about the impor-
tance of achieving euthyroidism before conception because of the risk
of decreased fertility and miscarriage. (wseg = 23/23.0/6.9/18.0)
Revision:
A doctor should talk with any woman with low thyroid function. She
needs to know that, because of the risk of low fertility or miscar-
riage, she needs to reach good thyroid function before she tries to
get pregnant. (wseg = 38/19.0/67.3/8.5)
S U MM A RY
This exercise asked you to practice talking in terms of one doctor treating one
patient. Table A3-23 shows the wseg scores for our revisions.
Original Revision
w s e g w s e g
1 38 38.0 16.8 20.3 37 18.5 60.0 9.4
2 24 24.0 37.9 13.9 23 23 76.8 8.2
3 12 12.0 60.7 7.7 14 14.0 65.7 7.5
4 27 27.0 10.2 18.5 34 17.0 57.7 9.4
5 27 27.0 35.2 15.0 32 16.0 71.6 7.2
6 23 23.0 6.9 18.0 38 19.0 67.3 8.5
Average 25.2 25.2 28.0 15.6 29.7 17.9 66.5 8.4
Change 4.5 −7.3 38.6 −7.2
181
This passage talks about real-world lung damage, but the terms, limitation, obstruc-
tion, destruction, and responsiveness sound abstract. We replaced them with the
more concrete-sounding terms, blockage, damage and narrowing.
Revision:
Fifth, the airflow blockage with COPD has a few key causes. They
are small airway disease, lung tissue damage (emphysema), and,
in many cases, other airway narrowing (asthma). (wseg = 28/14.0/
56.6/8.8)
3. Maintenance of nocturnal euglycemia is extremely important
and is challenging, since most cases of severe hypoglycemia occur at
night. (wseg = 19/19.0/4.9/17.2)
The risk of a blood clot is a real-world problem. Quantifying the risk involves an
abstract math calculation.
Revision:
Hives is a synonym for urticaria. Giant hives is a synonym for angioedema.1 Hives
and giant hives are real-world problems. Getting a history and doing a physical are
real-world actions. Medical diagnosis involves abstract thought.
Revision:
The initial workup for hives or giant hives is a history and physi-
cal. This data is used to figure out the likely cause. (wseg = 23/11.5/
70.1/6.3)
S U MM A RY
This exercise asked you to practice using short words to describe the real world.
Table A3-24 shows the wseg scores for our revisions.
183
Original Revision
w s e g w s e g
1 29 29.0 11.1 18.9 31 10.3 38.0 10.5
2 30 30.0 15.6 18.5 28 14.0 56.6 8.8
3 19 19.0 4.9 17.2 24 12.0 60.7 7.7
4 15 15.0 22.4 13.8 18 18.0 52.2 10.4
5 17 17.0 15.4 15.3 16 16.0 53.1 9.8
6 18 18.0 0.0 20.2 23 11.5 70.1 6.3
Average 21.3 21.3 11.6 17.3 23.3 13.6 55.1 8.9
Change 2.0 −7.7 43.6 −8.4
Data, method, finding, level, trend, age pattern and mortality are abstract ideas. An
appendix is part of the report.
184
Revision:
In this report, we present the data, methods and key findings from
the Global Burden of Disease Study 2010. We discuss global death
rates, trends, and age patterns. (wseg = 28/14.0/59.6/8.4)
3. Importantly, these trials all examine the initiation of therapy
with vitamin K antagonists and use as a primary end point the per-
centage of time that a patient is within the therapeutic range during
the initial phase of treatment. (wseg = 38/38.0/21.3/19.7)
A disease complication might involve something you can see in the real world.
Otherwise, this sounds abstract.
Revision:
The risk of getting HIV is a real-world risk. What people think about that risk is
abstract.
Revision:
S UM M A RY
This exercise asked you think about real-world vs. abstract possession or connec-
tion. Table A3-25 shows the wseg scores for our revisions.
Original Revision
w s e g w s e g
1 20 20.0 25.7 14.6 19 9.5 59.1 7.3
2 32 32.0 28.9 17.1 28 14.0 59.6 8.4
3 38 38.0 21.3 19.7 32 16.0 61.0 8.7
4 13 13.0 0.0 17.6 18 18.0 52.2 10.4
5 18 18.0 9.9 16.3 20 20.0 51.5 11.0
6 35 35.0 31.1 17.6 44 14.6 63.1 8.0
Average 26.0 26.0 19.5 17.2 26.8 15.4 57.8 9.0
Change 0.8 −10.7 38.3 −8.2
Revision:
Revision:
have recently been licensed in the United States and Europe, and
the results of several promising large phase III studies have been
recently published, now is an opportune time to review the current
treatment landscape for HCV, and to anticipate how that landscape
might look in coming years. (wseg = 142/35.5/24.8/18.6)
Revision:
Revision:
Revision:
S UM M A RY
This exercise asked you to practice using terms consistently. Table A3-26 shows
the wseg scores for our revisions.
188
Original Revision
w s e g w s e g
1 42 21.0 42.5 12.5 38 12.6 62.6 7.6
2 54 54.0 23.5 23.3 54 18.0 56.9 9.7
3 40 40.0 0.0 25.0 39 13.0 52.6 9.1
4 142 35.5 24.8 18.6 130 16.2 62.1 8.6
5 29 14.5 40.4 11.2 15 15.0 56.2 9.1
6 29 14.5 43.3 10.8 24 12.0 39.5 10.7
Average 56 29.9 29.1 16.9 50.0 14.5 55.0 9.1
Change −6.0 −15.5 25.9 −7.8
This sentence talks about real-world patients who could change their brain activity.
Revision:
This sentence uses the word regulatory in its common, non-medical sense, which
is abstract.
Revision:
We need a regulatory system that is more open. This could help fix
some of the drug industry’s poor behavior. (wseg = 20/10.0/65.5/6.6)
5. Some experimental support exists for the concept that the abil-
ity to discriminate between “self” and “nonself” involves learning to
respond aggressively when there are signals that suggest the pres-
ence of invasive pathogens and having effective regulatory mecha-
nisms for suppressing inflammatory responses when such signals
are absent. (wseg = 46/46.0/0.0/26.2)
S U MM A RY
This exercise asked you to practice talking about the real world using short,
concrete-sounding words. Table A3-27 shows the wseg scores for our revisions.
Original Revision
w s e g w s e g
1 26 26.0 27.5 15.8 27 13.5 58.3 8.4
2 19 19.0 13.8 16.0 23 11.5 70.1 6.3
3 16 16.0 31.9 12.7 16 16.0 68.9 7.6
4 17 17.0 0.4 17.4 20 10.0 65.5 6.6
5 46 46.0 0.0 26.2 41 13.6 71.2 6.7
6 23 23.0 0.0 20.0 23 11.5 59.0 7.8
Average 24.5 24.5 12.3 18.0 25.0 12.7 65.5 7.2
Change 0.5 −11.8 53.2 −10.8
• Elegant variation: (a) critical care specialists, ICU physicians, ICU clinicians,
health care providers; (b) an acceptable health state for the patient, an outcome that
patients can meaningfully appreciate; (c) critical care, intensive care, aggressive
critical care, intensive care interventions, care, therapeutic, treatment; and (d) spe-
cialists, physicians, clinicians, health care providers.
• Passive voice: should be considered, are often perceived.
• Abstract language: The many long words tend to sound abstract. The narrative
uses many plural subjects and objects, which tend to sound more abstract than
those in the singular.
• Formality: The long words also tend to sound formal.
• Nominalization: admission, transition, interventions, providers.
Revision:
Revision:
We think the shorter paragraphs and headings help a reader to scan the article
and get the “big picture” quickly before reading the details.
Revision:
We think the shorter paragraphs and headings help make the content easier
to grasp.
Revision:
Depression and anxiety did not increase in either group. But the
researchers cautioned, their findings might not apply to a smoker
whose depression is not treated with success. (wseg = 80/13.5/55.8/
8.7, excludes table)
2. What about health in Scotland? According to the UK’s national
statistical office, healthy life expectancy was 59·8 years for men
and 64·1 years for women in Scotland during 2008–10, 4·6 years
and 2·3 years fewer than for men and women in England, respec-
tively. According to the British Heart Foundation, 35% of Scottish
men and 30% of women have high blood pressure; alcohol use is one
noticeable contributor to ill health in Scotland, with up to 50% of
196
Revision:
Men Women
Healthy life expectancy (years)1 59.8 64.1
Difference compared to England (years) 1
(4.6) (2.3)
High blood pressure2 35% 30%
Alcohol use exceeding guidelines 2
50% 30%
Sources:
1. UK national statistical office; data for 2008–10
2. British Heart Foundation
70
60
50
40
30
20
10
0
Men Women
England Scotland
Women
Men
0 10 20 30 40 50 60
Figure A3-2 Scotland Health. 2
2. The narrative doesn’t explain this. Therefore, it appears to have left out a step
of reasoning. (The article does show this later in a figure.)
Note
1.
Stedman’s Medical Dictionary, s.vv. “Urticaria,” “Angioedema.”
200
201
GLOSSARY
The 1066 Principle—the general tendency for English speakers to use short words to talk about the
real world, and long words, more sparingly, to talk about abstract ideas.
Abstract—a theoretical way of looking at things; something that exists only in idealized form.
A term is abstract if it relates to the world of ideas, including a concept, theory, calculation or
procedure. Contrast with concrete.
Active voice—A sentence is in active voice when its subject is doing the action. See voice.
Clear—Writing is clear when the narrative uses words and concepts familiar to the reader. Ideally,
a reader can understand and vividly imagine the article on first reading without having to study it.
The reader remembers each key idea.
Closed compound—a compound word written as one word (e.g., multicell, hyperadrenergic, vasomotor,
and sinoatrial). See compound word.
Compound word or compound—a word formed by combining two or more words, or a word plus a
prefix. The three main types of compounds are the open compound (e.g., student nurse), the closed
compound (e.g., multicell), and the hyphenated compound (e.g., pre-menstrual).
Concise—Writing is concise when it demands as little of the reader’s mental energy as possible.
This usually means short while still clear. Good writing involves tradeoffs. A few short words may
convey the message more vividly than one long, but lifeless word. Writing concisely often means
cutting any unnecessary word; but sometimes, cutting too many words makes the message cryptic
and harder to understand.
Concrete—something from the real world (e.g., a doctor, a patient, a bed, a test tube). Contrast with
abstract.
Elegant variation—varying terms to make writing more interesting. Technical writing tends to
avoid elegant variation, but it is common in other types of writing.
Essential scientific content—important scientific ideas an author must include in their article.
Essential scientific term—a long word that helps convey essential scientific content clearly and
concisely. An essential scientific term meets four tests:
1. No shorter word serves just as well,
2. You can’t paraphrase in a few short words,
3. Doctors and other medical scientists use the term consistently (i.e., exclusively), and
4. It’s easy to look up in a standard reference.
False signal—using an abstract-sounding word to talk about something in the real world, or a
concrete-sounding word to talk about something abstract. Contrast with signal.
201
202
202 Glossary
Flesch Reading Ease—a readability test that indicates how difficult it is to read a passage in English.
The scores generally range from 0.0 to 100.0.
Flesch-Kincaid Grade Level—a readability test that assigns a USA school grade level or year to a
passage in English.
Hyphenated compound—a compound word where words are written together but separated by a
hyphen (e.g., pre-menstrual, cost-effective, one-time, self-reported). See compound word.
Insider—somebody who knows the science and vocabulary of a particular specialized field. Insiders
are the narrowest possible definition of the potential audience for an article. Contrast with the
widest reasonable audience.
Long word—any word with three or more syllables, but not including a two-syllable word that
becomes a three-syllable word by adding a common ending, such as -ed, -es or -ing.
Medicus incomprehensibilis—a condition that affects doctors and other medical scientists and causes
them to write dull, lifeless prose that is hard to understand. Medicus incomprehensibilis is primarily
caused by needless grammatical complexity.
Narrative pathway—the direction of a narrative, or a conceptual program for organizing a narrative.
Nominalization—the process of making an abstract noun out of a verb or adjective.
Noun string—a group of nouns and their modifiers. Often, a noun string consists of obscure
technical terms. Multiple terms may function together as an adjective.
Open compound—a compound word, where words work together, but are written as separate words
(e.g., student nurse, 50 percent, reference book). See compound word.
Passive voice—A sentence is in passive voice when its subject receives the action. See voice.
Past participle—a form of the verb that expresses completed action.1 A past participle is usually the
same form as the verb in past tense. For regular verbs, this means adding a -d or -ed ending (e.g.,
worked, decided, starved). Irregular verbs use irregular forms (e.g., broken, swum).2 Examples:
• “The results of the meta-analysis of treatment effect of lubiprostone vs. placebo are shown in
Figures 2 and 3.”3 In this sentence, shown is a past participle.
• “It may be specified in the protocol of a prospective accuracy study, for instance, that to reduce
study costs or burden to patients only a randomly selected subset of patients in a specific
subgroup are to be verified by the preferred reference standard.”4 In this sentence, specified,
selected, verified, and preferred are past participles.
Plain English—writing that conveys the right content, clearly and concisely. Writing in plain
English involves sharpening up the medical science to make it clearer and more accessible to the
widest reasonable audience.
Short word—a one-or two-syllable word. This also includes a two-syllable word that becomes a
three-syllable word by adding a common ending, such as -ed, -es, or -ing. Contrast with long word.
Signal—a way of indicating, through word choice, whether you’re talking about the real world
or the world of abstract ideas. Short words tend to signal real world, and longer words abstract.
Contrast with false signal.
Subject (grammar)—the noun or pronoun that agrees with the verb.5 A noun functioning as a
subject is the actor, person, or thing about which an assertion is made in a clause.6 Examples:
• “Onychomycosis is a fungal infection of the nails that causes discoloration, thickening, and
separation from the nail bed.”7 In this sentence, Onychomycosis is the subject.
• “Identification of hyphae, pseudohyphae, or spores confirms infection but does not identify the
organism.” 8 Here, Identification is the subject. (The phrase, “Identification of hyphae, pseudohyphae,
or spores,” is the logical subject.9)
203
Glossary 203
Notes
1. Chicago Manual of Style, 15th ed. §5.103.
2. Williams, Style: Lessons in Clarity, 266 (see Preface, n. 8).
3. Li F, et al. “Lubiprostone is Effective in the Treatment of Chronic Idiopathic Constipation and
Irritable Bowel Syndrome: A Systematic Review and Meta-Analysis of Randomized Controlled
Trials,” Mayo Clinic Proc 91, no. 4 (2016): 461.
4. Naaktgeboren C, et al. “Anticipating Missing Reference Standard Data When Planning
Diagnostic Accuracy Studies,” BMJ 352 (2016), under “The problem: missing reference stan-
dard data,” http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4772780/.
5. Cutts, Oxford Guide to Plain English, 122 (see chap. 1, n. 6).
6. The Chicago Manual of Style, 15th ed. §5.23.
7. Westerberg D, Voyack M, “Onychomycosis: Current Trends in Diagnosis and Treatment,” Am
Fam Phys 88, no. 11 (2013): 762.
8. Ibid.
9. See Williams, Style: Lessons in Clarity, 81–82 (see Preface, n. 8).
10. Merriam-Webster’s Learner’s Dictionary, Merriam- Webster.com, (accessed July 11, 2016),
http://www.merriam-webster.com/dictionary/verb. s.v. “Verb.”
11. Naaktgeboren et al. “Anticipating Missing Reference Standard,” (see Glossary, n. 4).
12. Palmer B, Clegg D, “Achieving the Benefits of High-Potassium, Paleolithic Diet, Without the
Toxicity,” Mayo Clinic Proc 91, no. 4 (2016): 500.
204
205
RESOURCES
Books
Booth, Wayne C., Gregory G. Colomb, Joseph M. Williams. The Craft of Research. 3rd ed. Chicago:
University of Chicago Press, 2008.
Cutts, Martin. Oxford Guide to Plain English. Oxford: Oxford University Press, 2009.
Garner, Bryan A. The Elements of Legal Style. 2nd ed. Oxford: Oxford University Press, 2002.
Garner, Bryan A. Legal Writing in Plain English. Chicago: University of Chicago Press, 2001.
Garner, Bryan A. Securities Disclosure in Plain English. Chicago: CCH Incorporated, 1999.
Greene, Anne E. Writing Science in Plain English. Chicago: University of Chicago Press, 2013.
Iverson, Cheryl, et al. AMA Manual of Style. 10th ed. Oxford: Oxford University Press, 2007.
Kimble, Joseph. Lifting the Fog of Legalese. Durham, NC: Carolina Academic Press, 2006.
Office of Investor Education and Assistance. A Plain English Handbook. Washington, DC: US
Securities and Exchange Commission, 1998.
Strunk, William, Jr. and E. B. White. The Elements of Style. 3rd ed. New York: Macmillan, 1979.
Stedman’s Medical Dictionary. 28th ed. Philadelphia: Lippincott Williams & Wilkins, 2006.
Tufte, Edward R. The Visual Display of Quantitative Information. 2nd ed. Cheshire, CT: Graphics
Press, 2001.
University of Chicago Press. Chicago Manual of Style. 15th ed. Chicago: University of Chicago
Press, 2003.
Williams, Joseph M. Style: Lessons in Clarity and Grace. 9th ed. New York: Pearson Longman, 2007.
Wydick, Richard C. Plain English for Lawyers. Durham, NC: Carolina Academic, 2005.
Articles
Gopen, George D., and Judith A. Swan. “The Science of Scientific Writing.” American Scientist 78
(November–December 1990), http://www.americanscientist.org/issues/pub/the-science-
of-scientific-writing/1.
Orwell, George. “Politics and the English Language,” Horizon (April 1946). Available online,
http://www.orwell.ru/library/essays/politics/english/e_polit/.
Internet References
Plain Language Action and Information Network. Federal Plain Language Guidelines. (March
2011, revised May 2011), www.plainlanguage.gov.
Plain Language at NIH, National Institutes of Health, https://www.nih.gov/institutes-nih/nih-
office-director/office-communications-public-liaison/clear-communication/plain-language.
205
206
╇ 207
INDEX
207
208
208 Index
Index 209
210 Index